Analysing how information flows along the layers of a multilayer perceptron is a topic of paramount importance in the field of artificial neural networks. After framing the problem from the point of view of information theory, in this position article a specific investigation is conducted on the way information is processed, with particular reference to the requirements imposed by supervised learning. To this end, the concept of information matrix is devised and then used as formal framework for understanding the aetiology of optimisation strategies and for studying the information flow. The underlying research for this article has also produced several key outcomes: i) the definition of a parametric optimisation strategy, ii) the finding that the optimisation strategy proposed in the information bottleneck framework shares strong similarities with the one derived from the information matrix, and iii) the insight that a multilayer perceptron serves as a kind of "adaptor", meant to process the input according to the given objective.

Using a martingale representation, we introduce a novel deep-learning approach, which we call DeepMartingale, to study the duality of discrete-monitoring optimal stopping problems in continuous time. This approach provides a tight upper bound for the primal value function, even in high-dimensional settings. We prove that the upper bound derived from DeepMartingale converges under very mild assumptions. Even more importantly, we establish the expressivity of DeepMartingale: it approximates the true value function within any prescribed accuracy $\varepsilon$ under our architectural design of neural networks whose size is bounded by $\tilde{c}\,D^{\tilde{q}}\varepsilon^{-\tilde{r}}$, where the constants $\tilde{c}, \tilde{q}, \tilde{r}$ are independent of the dimension $D$ and the accuracy $\varepsilon$. This guarantees that DeepMartingale does not suffer from the curse of dimensionality. Numerical experiments demonstrate the practical effectiveness of DeepMartingale, confirming its convergence, expressivity, and stability.

The Pontryagin-Thom theorem gives an isomorphism between the cobordism group of framed $n$-dimensional manifolds, $\omega_n$, and the $n^{th}$ stable homotopy group of the sphere spectrum, $\pi_n(\mathbb{S})$. The equivariant analogue of this theorem, gives an isomorphism between the equivariant cobordism group of $V$-framed $G$-manifolds, $\omega_V^G$, and the $V^{th}$ equivariant stable homotopy group of the $G$-sphere spectrum, $\pi_V^G(\mathbb{S})$, for a finite group $G$ and a $G$-representation, $V$. In this paper, we explicitly identify the images of each element of $\omega_1^{C_2}$ and $\omega_\sigma^{C_2}$ in $\pi_1^{C_2}(\mathbb{S})$ and $\pi_\sigma^{C_2}(\mathbb{S})$ under the equivariant Pontryagin-Thom isomorphism.

Modern FFT/NTT analytics, coded computation, and privacy-preserving ML interface routinely move polynomial frames across NICs, storage, and accelerators. However, even rare silent data corruption (SDC) can flip a few ring coefficients and cascade through downstream arithmetic. Conventional defenses are ill-matched to current low-latency pipelines: detect-and-retransmit adds RTTs, while byte-stream ECC ignores the algebraic structure and forces format conversions. To that end, we propose a structure-preserving reliability layer that operates in the encoded data's original polynomial ring, adds a small amount of systematic redundancy, and corrects symbol errors/flagged erasures without round-trip or format changes. We construct two complementary schemes: one for odd length $N_{odd}$ via a Hensel-lifted BCH ideal with an idempotent encoder, and one for power-of-two length $N_{2^m}$ via a repeated-root negacyclic code with derivative-style decoding. In particular, to stay robust against clustered errors, a ring automorphism provides in-place interleaving to disperse bursts. Implementation wise, on four frame sizes $N\!=\!1024, 2048, 4096, 8192$, we meet a per-frame failure target of $10^{-9}$ at symbol error rates $10^{-6}\text{--}10^{-5}$ with $t\!=\!8\text{--}9$, incurring only $0.20\%\text{--}1.56\%$ overhead and tolerating $\sim\!32\text{--}72$\,B unknown-error bursts (roughly doubled when flagged as erasures) after interleaving. By aligning error correction with ring semantics, we take a practical step toward deployable robustness for polynomial-frame computations from an algebraic coding perspective.

In this work, we investigate Maker-Breaker directed triangle games -- a directionally constrained variant of the classical Maker-Breaker triangle game. Our board of interest is a tournament, and the winning sets constitute all directed triangles ($3$-cycles) present in the tournament. We begin by studying the Maker-Breaker directed triangle game played on a specially defined tournament called the parity tournament, and we identify the board size threshold to be $n=7$, which is to say that if the size (i.e.\ the number of vertices) of the parity tournament equals $n$, Breaker has a winning strategy for $3\leq n< 7$, while Maker can ensure a win for herself for $n\geq 7$. For the $(1:b)$ biased version of this game, we prove that the bias threshold $b^*(n)$ satisfies $\sqrt{\left(1/12+o(1)\right)\ n}\leq b^{*}(n) \leq\sqrt{\left(8/3+o(1)\right)\ n}$, which matches the order of magnitude (\ $\sqrt{n}$) of the bias threshold for the undirected counterpart of this game. Next, we consider the game on random tournaments $T(n,p)$ with labeled vertices $1,2,\ldots,n$, such that the edge between $i$ and $j$, for each $i<j$, is directed from $i$ towards $j$ with probability $p$, independent of all else. We prove that Maker wins this game with probability approaching $1$ as $n \to \infty$ for any fixed $p \in (0,1)$. Extending the notion of `bias' from undirected games to our directed framework, we introduce the $(1: \kappa(n))$ flip-biased Maker-Breaker directed triangle game on the parity tournament, where we allow Breaker to strategically flip the directions of a fixed number of edges $\kappa(n)$ before the game begins. We show that the flip-bias threshold $\kappa^*(n)$ is asymptotically equal to $n^2/4$ as $n \to \infty$. This work opens up the possibility of studying a variety of directionally constrained Maker-Breaker positional games on tournaments (and more generally, on directed graphs).

In this paper we determine the congruence of Jacobi sums $J(1, 1)_{49}$ of order 49 over a field $\mathbb{F}_p$. We also show that simpler congruences hold for $J(1, 1)_{49}$ in the case of artiad and hyperartiad primes.

We prove that the Hilbert property is satisfied by certain del Pezzo surfaces of degree one and Picard rank 1 over fields finitely generated over $\mathbb{Q}$. We generalize results of the first author on elliptic surfaces and employ constructions used by Desjardins and the third author to prove density of rational points. Our results are the first on the Hilbert property for minimal del Pezzo surfaces of degree one without a conic fibration.

Given a set $X$ and a collection ${\mathcal H}$ of functions from $X$ to $\{0,1\}$, the VC-dimension measures the complexity of the hypothesis class $\mathcal{H}$ in the context of PAC learning. In recent years, this has been connected to geometric configuration problems in vector spaces over finite fields. In particular, it is easy to show that the VC-dimension of the set of spheres of a given radius in $\mathbb{F}_q^d$ is equal to $d+1$, since this is how many points generically determine a sphere. It is known that for $E\subseteq \mathbb{F}_q^d$, $|E|\geq q^{d-\frac{1}{d-1}}$, the set of spheres centered at points in $E$, and intersected with the set $E$, has VC-dimension either $d$ or $d+1$. In this paper, we study a similar question over Euclidean space. We find an explicit dimensional threshold $s_d<d$ so that whenever $E\subseteq \mathbb{R}^d$, $d\geq 3$, and the Hausdorff dimension of $E$ is at least $s_d$, it follows that there exists an interval $I$ such that for any $t\in I$, the VC-dimension of the set of spheres of radius $t$ centered at points in $E$, and intersected with $E$, is at least $3$. In the process of proving this theorem, we also provide the first explicit dimensional threshold for a set $E\subseteq \mathbb{R}^3$ to contain a $4$-cycle, i.e. $x_1,x_2,x_3,x_4\in E$ satisfying $$ |x_1-x_2|=|x_2-x_3|=|x_3-x_4|=|x_4-x_1| $$

Consider $F$ an element of the $p$-th Wiener chaos $\WW_p$, and denote by $\prob_F$ its law. For a positive integer $m$, let $\boldsymbol{\gamma}_{F,m}$ be the Radon measure with density $x \mapsto \frac{e^{-x^2/2}}{\sqrt{2\pi}} \left(1 + \sum_{k=3}^{4m-1} \frac{\E[H_k(F)]}{k!}\, H_k(x)\right)$, where $H_k$ is the $k$-th Hermite polynomial. The main goal of this article is to prove that the total variation distance between $\prob_F$ and $\boldsymbol{\gamma}_{F,m}$ is of order $\Var(\Gamma(F,F))^{({m+1})/{2}}$, where $\Gamma(F,F)$ denotes the carré-du-champ operator of $F$. The variance of $\Gamma(F,F)$ is known to govern Gaussian fluctuations and can be bounded from above by $\kappa_4(F)$, the fourth cumulant of $F$, as established in the seminal work \cite{NP2009a}. Our result thus provides a genuine Edgeworth expansion in the setting of central convergence on Wiener chaoses. In this context, the quantity $\Var(\Gamma(F,F))$ plays the role of the small parameter that governs the accuracy of the approximation, in the same way that $1/\sqrt{n}$ does in the classical central limit theorem. To the best of our knowledge, our work is the first to establish Edgeworth expansions for Wiener chaoses in full generality and at arbitrary order, together with explicit remainder bounds that systematically improve with the order of the expansion--exactly as one would expect from an Edgeworth approximation. Our results apply verbatim to every situation where a central limit theorem is available for chaos elements, since no structural assumption is required beyond belonging to a fixed Wiener chaos. As a byproduct, we recover the celebrated optimal fourth moment theorem from \cite{NP2015} by combining the expansions at the first and second orders, with sharper quantitative bounds. Previous works on Edgeworth expansions for Wiener chaoses were essentially restricted to the first order.

The problem of optimal estimation of linear functionals $A {\xi}=\int_{0}^{\infty} a(t)\xi(t)dt$ and $A_T{\xi}=\int_{0}^{T} a(t)\xi(t)dt$ depending on the unknown values of random process $\xi(t)$, $t\in R$, with stationary $n$th increments from observations of ttis process for $t<0$ is considered. Formulas for calculating mean square error and spectral characteristic of optimal linear estimation of the functionals are proposed in the case when spectral density is exactly known. Formulas that determine the least favorable spectral densities are proposed for given sets of admissible spectral densities.

The Cauchy dual subnormality problem (for short, CDSP) asks whether the Cauchy dual of a $2-$isometry is subnormal. In this article, we prove that if $\mu$ is a linear combination of unit point mass measures at two non-antipodal points on the unit circle, then the Cauchy dual $M_z'$ of the multiplication operator $M_z$ on the Dirichlet space $D(\mu)$ is not subnormal. If the two points are antipodal then the subnormality of the said operator has been already established in the literature. Thus, we have a complete solution of CDSP in this case.

Suppose $K$ is an imaginary quadratic field, and let $N_K$ denote the field norm in the ring of integers $O_K$. Let $B(x_0,r) = \{x \in O_K: |N_K(x-x_0)| < r\}$. Let $G_K(X) = \max \{r > 0: \text{there exists } x_0 \in O_K \text{ such that } |N_K(x_0)| \leq X \text{ and } B(x_0,r) \text{ contains no primes} \}$. We show that $ G_{K}(X) \gg_K (\log X) \frac{\log_2(X) \log_4(X)}{\log_3(X)}$.

In this work, we establish connections between the theory of algebraic $n$-valued monoids and groups and the theories of discriminants and projective duality. We show that the composition of projective duality followed by the Möbius transformation $z\mapsto 1/z$ defines a shift operation $\mathbb{M}_n(\mathbb{C}P^1)\mapsto \mathbb{M}_{n-1}(\mathbb{C}P^1)$ in the family of algebraic $n$-valued coset monoids $\{\mathbb{M}_{n}(\mathbb{C}P^1)\}_{n\in\mathbb{N}}$. We also show that projective duality sends each Fermat curve $x^n+y^n=z^n$ $(n\ge 2)$ to the curve $p_{n-1}(z^n; x^n, y^n)=0$, where the polynomial $p_n(z;x,y)$ defines the addition law in the monoid $\mathbb{M}_n(\mathbb{C}P^1)$. We solve the problem of describing coset $n$-valued addition laws constructed from cubic curves. As a corollary, we obtain that all such addition laws are given by polynomials, whereas the addition laws of formal groups on general cubic curves are given by series.

Kidney Exchange Programs (KEPs) promote access to living donor trans- plantation for patients suffering from end stage renal disease. The HLA compatibility between transplant recipients and donors plays an important role when solving the matching problems occuring in KEPs. Compatibility defines the feasible solution space and often occurs in a weighted form in the objective function. Recently, the paradigms used to express HLA com- patibility have advanced substantially from antigens to alleles and eplets. We show how KEP effectiveness and equity vary with these three paradigms. The equity analysis focuses on ethnic inequities and the result confirm that moving from antigen to allele and eplet paradigms may enlarge inequities. We present new optimization models based on equity weighting that rem- edy ethnic inequities for all three paradigms without sacrificing access to transplant.

We develop a variant of Stein's method of comparison of generators to bound the Kolmogorov, total variation, and Wasserstein-1 distances between distributions on the real line. Our discrepancy is expressed in terms of the ratio of reverse hazard rates; it therefore remains tractable even when density derivatives are intractable. Our main application concerns the approximation of normalized extremes by Fréchet laws. In this setting, the new discrepancy provides a quantitative measure of distributional proximity in terms of the average regular variation at infinity of the underlying cumulative distribution function. We illustrate the approach through explicit computations for maxima of Pareto, Cauchy, and Burr~XII distributions.

We prove that a three dimensional compact Ricci flow that encounters a Type I singularity has uniformly bounded diameter up to the singular time, thus giving an affirmative answer - for Type I singularities - to a conjecture of Perelman. To achieve this, we introduce a concept of a neck-region for a Ricci flow, analogous to the neck-regions introduced by Jiang-Naber and Cheeger-Jiang-Naber, in the study of Ricci limit spaces. We then prove that the associated packing measure is, in a certain sense, Ahlfors regular, a result that holds in any dimension.

The purpose of this note is to prove the celebrated Discrete Renewal Theorem in a common special case. We use only very elementary methods from real analysis, rather than markov chain theory, complex analysis, or generating functions. Provided is an introduction to a 1d discrete renewal process via a board game example, our proof the discrete renewal theorem, as well as background and history of the Erdǒs-Feller-Pollard Theorem.

This paper deals with the problem of optimal mean-square filtering of the linear functionals $A{\xi}=\int_{0}^{\infty}a(t)\xi(-t)dt$ and $A_T{\xi}=\int_{0}^Ta(t)\xi(-t)dt$ which depend on the unknown values of random process $\xi(t)$ with stationary $n$th increments from observations of process $\xi(t)+\eta(t)$ at points $t\leq0$, where $\eta(t)$ is a stationary process uncorrelated with $\xi(t)$. We propose the values of mean-square errors and spectral characteristics of optimal linear estimates of the functionals when spectral densities of the processes are known. In the case where we can operate only with a set of admissible spectral densities relations that determine the least favorable spectral densities and the minimax spectral characteristics are proposed.

The representation theory of tensor functions is a powerful mathematical tool for constitutive modeling of anisotropic materials. A major limitation of the traditional theory is that many point groups require fourth- or sixth-order structural tensors, which significantly impedes practical engineering applications. Recent advances have introduced a reformulated representation theory that enables the modeling of anisotropic materials using only lower-order structural tensors (i.e., second-order or lower). Building upon the reformulated theory, this work establishes the representations of tensor functions for three-dimensional centrosymmetric point groups. For each point group, we propose a lower-order structural tensor set and derive the representations of tensor functions explicitly. For scalar-valued and second-order symmetric tensor-valued functions, our theory is indeed applicable to all three-dimensional point groups because their representations are determined by the corresponding centrosymmetric groups. The representation theory presented here is broadly applicable for constitutive modeling of anisotropic materials.

This paper introduces and systematically develops the theory of polyadic group rings, a higher arity generalization of classical group rings $\mathcal{R}[\mathsf{G}]$. We construct the fundamental operations of these structures, defining the $\mathbf{m}_{r}$-ary addition and $\mathbf{n}_{r} $-ary multiplication for a polyadic group ring $\mathrm{R}^{[\mathbf{m} _{r},\mathbf{n}_{r}]}=\mathcal{R}^{[m_{r},n_{r}]}[\mathsf{G}^{[n_{g}]}]$ built from an $(m_{r},n_{r})$-ring and an $n_{g}$-ary group. A central result is the derivation of the "quantization" conditions that interrelate these arities, governed by the arity freedom principle, which also extends to operations with higher polyadic powers. We establish key algebraic properties, including conditions for total associativity and the existence of a zero element and identity. The concepts of the polyadic augmentation map and augmentation ideal are generalized, providing a bridge to the classical theory. The framework is illustrated with explicit examples, solidifying the theoretical constructions. This work establishes a new foundation in ring theory with potential applications in cryptography and coding theory, as evidenced by recent schemes utilizing polyadic structures.

The problem of optimal linear estimation of functional depending on the unknown values of periodically correlated stochastic process from observations of this process for is considered. Formulas that determine the least favorable processes and the minimax estimation for functional are proposed for the given class of admissible processes.

A fully discrete Crank--Nicolson Leap--Frog (CNLF) scheme is proposed and analyzed for the unsteady bioconvection flow problem with concentration-dependent viscosity. Spatial discretization is handled via the Galerkin finite element method (FEM), while temporal discretization employs the CNLF method for the linear terms and a semi-implicit approach for the nonlinear terms. The scheme is proven to be unconditionally stable, i.e., the time step is not subject to a restrictive upper bound. Using the energy method, $L^2$-optimal error estimates are derived for the velocity and concentration . Finally, numerical experiments are presented to validate the theoretical results.

In the seminal article \cite{LED16}, an integral representation of the derivatives of entropy along the heat flow of a probability measure was established under suitable moment conditions. These integral representations have found significant applications in diverse domains - notably in information theory (e.g., entropy power inequalities, monotonicity of Fisher information) and in estimation theory (through the link between entropy derivatives and the minimum mean square error, MMSE, in Gaussian channels). The representations involve multivariate polynomials $(R_n)_n$, arising from a Lie algebra framework on multilinear operators. Despite their central role, the combinatorial structure of these polynomials remains only partially understood. In this note, we prove that the number of monomials in $R_n$ coincides with the number of degree sequences with degree sum $2n$ having a non-separable graph realization, thereby resolving a conjecture from \cite{MPS24}, and drawing an interesting link between these two domains.

We revisit an ingenious argument of K. Ball to provide sharp estimates for the volume of sections of a convex body in John's position. Our technique combines the geometric Brascamp-Lieb inequality with a generalised Parseval-type identity. This lets us complement some earlier results of the first two named authors, as well as generalise the classical estimates of Meyer-Pajor and Koldobsky regarding extremal sections of $B_p^n$ balls to a broader family of norms induced by a John's decomposition of the identity in $\mathbb{R}^n$.

Reliable inference from complex survey samples can be derailed by outliers and high-leverage observations induced by unequal inclusion probabilities and calibration. We develop a minimum Hellinger distance estimator (MHDE) for parametric superpopulation models under complex designs, including Poisson PPS and fixed-size SRS/PPS without replacement, with possibly stochastic post-stratified or calibrated weights. Using a Horvitz-Thompson-adjusted kernel density plug-in, we show: (i) $L^1$-consistency of the KDE with explicit large-deviation tail bounds driven by a variance-adaptive effective sample size; (ii) uniform exponential bounds for the Hellinger affinity that yield MHDE consistency under mild identifiability; (iii) an asymptotic Normal distribution for the MHDE with covariance $\mathbf A^{-1}\boldsymbol\Sigma \mathbf A^{\intercal}$ (and a finite-population correction under without-replacement designs); and (iv) robustness via the influence function and $\alpha$-influence curves in the Hellinger topology. Simulations under Gamma and lognormal superpopulation models quantify efficiency-robustness trade-offs relative to weighted MLE under independent and high-leverage contamination. An application to NHANES 2021-2023 total water consumption shows that the MHDE remains stable despite extreme responses that markedly bias the MLE. The estimator is simple to implement via quadrature over a fixed grid and is extensible to other divergence families.

We prove that (local) input-to-state stability ((L)ISS) and integral input-to-state stability (iISS) of time-varying infinite-dimensional systems in abstract spaces follows from the existence of a {corresponding} Lyapunov function. In particular, input-to-state stability of linear time-varying control systems in Hilbert spaces with bounded input operators is discussed. Methods for the construction of non-coercive LISS/iISS Lyapunov functions are presented for a certain class of time-varying semi-linear evolution equations. Two examples are given to illustrate the effectiveness of the results.

Given a number field $K$, we completely classify the preperiodic portraits of the maps $x^d+c$ where $c\in K$ is an algebraic integer and $d$ is sufficiently large depending on the degree of $K$. Specifically, we show that there are exactly thirteen such portraits up to the natural action of roots of unity. In particular, we obtain some of the main results of recent work of the authors unconditionally for algebraic integers by replacing the use of the abc-conjecture with bounds on linear forms in logarithms. We then include applications of this work to several problems in semigroup dynamics, including the construction of irreducible polynomials and the classification of post-critically finite sets.

We propose a new radiation condition for an infinite inhomogeneous two-dimensional medium which is periodic in the vertical direction and remains invariant in the horizontal direction. The classical Rayleigh-expansion radiation condition does not apply to our case, because this would require the medium to be inhomogeneous in a half plane. We utilize the Floquet theory to derive upward/downward wave modes and define radiation conditions by expansions w.r.t. these modes. The downward radiation conditions leads to a downward Dirichlet-to-Neumann map which can be used to truncate the infinite inhomogeneous domain in the vertical direction. So we prove mapping properties of the upward/downward Dirichlet-to-Neumann maps based on the asymptotic behavior of high-order wave modes. Finally, we verify the strong ellipticity of the sesquilinear form corresponding to the new scattering problem and show the unique solvability for all wavenumbers with the exception of a countable set of numbers bounded below by a small positive constant.

In this work, we introduce a novel computational framework for solving the two-dimensional Hele-Shaw free boundary problem with surface tension. The moving boundary is represented by point clouds, eliminating the need for a global parameterization. Our approach leverages Generalized Moving Least Squares (GMLS) to construct local geometric charts, enabling high-order approximations of geometric quantities such as curvature directly from the point cloud data. This local parameterization is systematically employed to discretize the governing boundary integral equation, including an analytical formula of the singular integrals. We provide a rigorous convergence analysis for the proposed spatial discretization, establishing consistency and stability under certain conditions. The resulting error bound is derived in terms of the size of the uniformly sampled point cloud data on the moving boundary, the smoothness of the boundary, and the order of the numerical quadrature rule. Numerical experiments confirm the theoretical findings, demonstrating high-order spatial convergence and the expected temporal convergence rates. The method's effectiveness is further illustrated through simulations of complex initial shapes, which correctly evolve towards circular equilibrium states under the influence of surface tension.

In this article, we investigate the theory of weighted functions of bounded variation (BV), as introduced by Baldi [Ba01]. Depending on the theorem, we impose lower semicontinuity and/or a pointwise A1 condition on the weight. Our motivation is twofold: to establish weighted Gagliardo-Nirenberg-Sobolev (GNS) inequalities for BV functions, and to clarify and extend earlier results on weighted BV spaces. Our main contributions include a structure theorem under minimal assumptions (lower semicontinuity), a smooth approximation result, an embedding theorem, a weighted GNS inequality for BV functions, and a corresponding weighted isoperimetric inequality.

In this paper we consider $\phi$-mixing measures and show that the limiting return times distribution is compound Poisson distribution as the target sets shrink to a zero measure set. The approach we use generalises a method given by Galves and Schmitt in 1997 for the first entry time to higher orders.

Many real-world decision-making processes rely on solving mixed-integer nonlinear programming (MINLP) problems. However, finding high-quality solutions to MINLPs is often computationally demanding. This has motivated the development of specialized algorithms that take advantage of the structure of some MINLPs to improve their tractability. In this work, we propose the Mixed-integer Smoothing Surrogate Optimization with Constraints (MiSSOC) approach, a novel optimization algorithm that builds and solves surrogate problems for complex MINLPs which are more tractable in practice. MiSSOC integrates statistical modeling into mathematical optimization by approximating complex functions in an MINLP using smooth additive regression models with $B-$splines. Expert knowledge can be incorporated into the building phase of the approximating functions through shape constraints related to sign (or bounds), monotonicity and convexity over the observed domain. Thus, MiSSOC fills a gap in the literature by building surrogates that are both data-driven and knowledge-driven. The surrogate problem is formulated by replacing the original complex functions with their approximations. The proposed MiSSOC algorithm is evaluated through a set of experiments, including benchmark instances and a real-world case study. MiSSOC is tested with different state-of-the-art solvers and with the tailored Sequential Convex MINLP (SC-MINLP) algorithm. The latter exploits the separable structure of the surrogate functions, which results from the approximating functions being sums of piecewise univariate polynomials. MiSSOC is benchmarked against solving the original MINLPs directly. The results show that MiSSOC is an effective approach for addressing complex MINLPs through surrogate modeling, particularly when used in combination with the SC-MINLP algorithm.

We investigate the large-time behavior of the pressureless Euler system with nonlocal velocity alignment and interaction forces, with the aim of characterizing the asymptotic convergence of classical solutions under general interaction potentials $W$ and communication weights. We establish quantitative convergence in three settings. In one dimension with $(\lambda,\Lambda)$-convex potentials, i.e., potentials satisfying uniform lower and upper quadratic bounds, bounded communication weights yield exponential decay, while weakly singular ones lead to sharp algebraic rates. For the Coulomb--quadratic potential $W(x)=-|x|+\frac12 |x|^2$, we prove exponential convergence for bounded communication weights and algebraic upper bounds for singular communication weights. In a multi-dimensional setting with uniformly $(\lambda,\Lambda)$-convex potentials, we show exponential decay for bounded weights and improved algebraic decay for singular ones. In all cases, the density converges (up to translation) to the minimizer of the interaction energy, while the velocity aligns to a uniform constant. A unifying feature is that the convergence rate depends only on the local behavior of communication weights: bounded kernels yield exponential convergence, while weakly singular ones produce algebraic rates. Our results thus provide a comprehensive description of the asymptotic behavior of Euler--alignment dynamics with general interaction potentials.

We establish an alternative, ``perpendicular" collection of generating functions for the coefficients of Gaussian polynomials, $\begin{bmatrix}N+m\\m\end{bmatrix}_q$. We provide a general characterization of these perpendicular generating functions. For small values of $m$, unimodality of the coefficients of Gaussian polynomials is easily proved from these generating functions. Additionally, we uncover new and surprising identities for the differences of Gaussian polynomial coefficients, including a very unexpected infinite family of congruences for coefficients of $\begin{bmatrix}N+4\\4\end{bmatrix}_q$.

Natural disasters threaten the resilience of power systems, causing widespread power outages that disrupt critical loads (e.g., hospitals) and endanger public safety. Compared to the conventional restoration methods that often have long response times, leveraging government-controlled electric school buses (ESBs) with large battery capacity and deployment readiness offers a promising solution for faster power restoration to critical loads during disasters while traditional maintenance is underway. Therefore, we study the problem of routing and scheduling a heterogeneous fleet of ESBs to satisfy the energy demand of critical isolated loads around disasters addressing the following practical aspects: combined transportation and energy scheduling of ESBs, multiple back-and-forth trips of ESBs between isolated loads and charging stations, and spatial-wise coupling among multiple ESB routes. We propose an efficient mixed-integer programming model for routing and scheduling ESBs, accounting for the practical aspects, to minimize the total restoration cost over a planning horizon. We develop an efficient exact branch-and-price (B&P) algorithm and a customized heuristic B&P algorithm integrating dynamic programming and labeling algorithms. Numerical results based on a real case study of San Antonio disaster shelters and critical facilities demonstrate that our proposed exact B&P and heuristic B&P algorithms are computationally 121 and 335 times faster, respectively, than Gurobi. Using network sparsity to incorporate the limitation in shelter-ESB type compatibility in the model demonstrates that the total restoration cost increases, on average, by 207% as the network becomes fully sparse compared to fully connected. The capacity utilization metric reflects that the proposed practical ESB routing and scheduling enables an ESB to meet the energy demand 4.5 times its effective usable capacity.

Mathematical modelling has traditionally relied on detailed system knowledge to construct mechanistic models. However, the advent of large-scale data collection and advances in machine learning have led to an increasing use of data-driven approaches. Recently, hybrid models have emerged that combine both paradigms: well-understood system components are modelled mechanistically, while unknown parts are inferred from data. Here, we focus on one such class: universal differential equations (UDEs), where neural networks are embedded within differential equations to approximate unknown dynamics. When fitted to data, these networks act as universal function approximators, learning missing functional components. In this work, we note that UDE identifiability, i.e. our ability to identify true system properties, can be split into parametric and functional identifiability (assessing identifiability for the mechanistic and data-driven model parts, respectively). Next, we investigate how UDE properties, such as neural network numbers and constraints, affect parametric and functional identifiability. Notably, we show that across a wide range of models, the generalisation of a fully mechanistic model to a UDE has little impact on the mechanistic components' parametric identifiability. Finally, we note that hybrid modelling through the fitting of unknown functions (as achieved by UDEs) is particularly well-suited to chemical reaction network (CRN) modelling. Here, CRNs are used in fields ranging from systems biology, chemistry, and pharmacology to epidemiology and population dynamics, making them highly relevant for study. By showcasing how CRN-based UDE models can be highly interpretable, we demonstrate that this hybrid approach is a promising avenue for future applications.

Saddle points provide a hierarchical view of the energy landscape, revealing transition pathways and interconnected basins of attraction, and offering insight into the global structure, metastability, and possible collective mechanisms of the underlying system. In this work, we propose a stochastic saddle-search algorithm to circumvent exact derivative and Hessian evaluations that have been used in implementing traditional and deterministic saddle dynamics. At each iteration, the algorithm uses a stochastic eigenvector-search method, based on a stochastic Hessian, to approximate the unstable directions, followed by a stochastic gradient update with reflections in the approximate unstable direction to advance toward the saddle point. We carry out rigorous numerical analysis to establish the almost sure convergence for the stochastic eigenvector search and local almost sure convergence with an $O(1/n)$ rate for the saddle search, and present a theoretical guarantee to ensure the high-probability identification of the saddle point when the initial point is sufficiently close. Numerical experiments, including the application to a neural network loss landscape and a Landau-de Gennes type model for nematic liquid crystal, demonstrate the practical applicability and the ability for escaping from "bad" areas of the algorithm.

This paper studies high-order partial differential equations with random initial conditions that have both long-memory and cyclic behavior. The cases of random initial conditions with the spectral singularities, both at zero (representing classical long-range dependence) and at non-zero frequencies (representing cyclic long-range dependence), are investigated. Using spectral methods and scaling techniques, it is proved that, after proper rescaling and normalization, the solutions converge to Gaussian random fields. For each type of equation, spectral representations and covariance functions of limit fields are given. For odd-order equations, we apply the kernel averaging of solutions to obtain nonexplosive and nondegenerate limits. It is shown that the different limit fields are determined by the even or odd orders of the equations and by the presence or absence of a spectral singularity at zero. Several numeric examples illustrate the obtained theoretical results.

In this paper, we present a unified framework for studying cohomology theories of various operators in the context of pseudoalgebras. The central tool in our approach is the notion of a quasi-twilled Lie pseudoalgebra. We introduce two types of deformation maps. Type I unifies modified $r$ matrices, crossed homomorphisms, derivations, and homomorphisms; and Type II provides a uniform treatment of relative Rota-Baxter operators, twisted Rota-Baxter operators, Reynolds operators, and deformation maps of matched pairs of Lie conformal algebras. We construct the corresponding controlling algebras and define cohomology theories for both types of deformation maps. These results recover existing cohomological results for known operators and yield new results, including the cohomology theory for modified $r$-matrices and deformation maps of matched pairs of Lie pseudoalgebras.

This is a free textbook suitable for a one-semester course on Markov chains, covering basics of finite-state chains, many classical models, asymptotic behavior and mixing times, Monte Carlo methods, and martingales and harmonic functions. It is designed to fill a gap in the literature by being suitable for undergraduates; much of the theory is thus built from the ground up, with only basic probability and linear algebra assumed. We take as our basic framework the first four chapters of the classic Levin-Peres text "Markov Chains and Mixing Times," generously expanding to make an exposition suitable for an undergraduate audience. We also incorporate over a hundred exercises and problems, along with a rich set of accompanying illustrations. Suggested homework sets are found in an appendix. Updated editions will periodically appear as new versions of this submission.

For each positive integer $n$, we denote by $\omega^*(n)$ the number of shifted-prime divisors $p-1$ of $n$, i.e., \[\omega^*(n):=\sum_{p-1\mid n}1.\] First introduced by Prachar in 1955, this function has interesting applications in primality testing and bears a strong connection with counting Carmichael numbers. Prachar showed that for a certain constant $c_0 > 0$, \[\omega^*(n)>\exp\left(c_0\frac{\log n}{(\log\log n)^2}\right)\] for infinitely many $n$. This result was later improved by Adleman, Pomerance and Rumely, who established an inequality of the same shape with $(\log\log n)^2$ replaced by $\log\log n$. Assuming the Generalized Riemann Hypothesis for Dirichlet $L$-functions, Prachar also proved the stronger inequality \[\omega^*(n)>\exp\left(\left(\frac{1}{2}\log2+o(1)\right)\frac{\log n}{\log\log n}\right)\] for infinitely many $n$. By refining the arguments of Prachar and of Adleman, Pomerance and Rumely, we improve on their results by establishing \begin{align*} \omega^*(n)&>\exp\left(0.6736\log 2\cdot\frac{\log n}{\log\log n}\right) \quad\text{(unconditionally)},\\ \omega^*(n)&>\exp\left(\left(\log\left(\frac{1+\sqrt{5}}{2}\right)+o(1)\right)\frac{\log n}{\log\log n}\right) \quad\text{(under GRH)}, \end{align*} for infinitely many $n$.

We discuss the following inverse problem: given the run-up data of a tsunami wave, can we recover its initial shape? We study this problem within the framework of the non-linear shallow water equations, a model widely used to study tsunami propagation and inundation. Previously, it has been demonstrated that in the case of infinite sloping bathymetry, it is possible to recover the initial water displacement and velocity from shoreline readings \cite{Rybkin23,Rybkin24,Rybkin25}. We consider a finite sloping bathymerty. We show that it is possible to recover boundary conditions (water displacement and velocity) on a virtual buoy from the shoreline data. Further, we discuss stitching together the shallow water equations and the Boussinesq equation in a more complex piece-wise sloping bathymetry in order to recover the initial conditions, while incorporating the dispersion to our model.

Let $G$ be a two-step nilpotent Lie group, identified via the exponential map with the Lie-algebra $\mathfrak g=\mathfrak g_1\oplus\mathfrak g_2$, where $[\mathfrak g,\mathfrak g]\subset \mathfrak g_2$. We consider maximal functions associated to spheres in a $d$-dimensional linear subspace $H$, dilated by the automorphic dilations. $L^p$ boundedness results for the case where $H=\mathfrak g_1$ are well understood. Here we consider the case of a tilted hyperplane $H\neq \mathfrak g_1$ which is not invariant under the automorphic dilations. In the case of Métivier groups it is known that the $L^p$-boundedness results are stable under a small linear tilt. We show that this is generally not the case for other two-step groups, and provide new necessary conditions for $L^p$ boundedness. We prove these results in a more general setting with tilted versions of submanifolds of $\mathfrak g_1$.

Let $\psi$ be a holomorphic function on the open unit ball $\BB \subset \C^N$, and let $\varphi$ be a holomorphic self-map of $\BB$, associated with normal weights $\nu$ and $\mu$. We consider the weighted composition operator $ W_{\psi,\varphi} : \mathcal H_\nu^{(n)} \to \mathcal H_\mu^{(m)}, \quad n,m \in \N,$ acting between weighted-type high-order growth spaces. Unlike previous studies that involve the full symbol $\varphi$, this paper establishes characterizations of the boundedness, compactness, and asymptotic norm estimates of $W_{\psi,\varphi}$ \emph{solely in terms of the symbol $\psi$ and a single component function $\varphi_p$ of $\varphi$}, offering a new approach to the analysis of such operators.

Superconvergent and divergence-free finite element methods for the Stokes equation are developed. The velocity and pressure are discretized using $H(\mathrm{div})$-conforming vector elements and discontinuous piecewise polynomials. The discrete formulation employs a weak deviatoric gradient operator built with tangential-normal continuous finite elements for traceless tensors, requiring no stabilization. Optimal and superconvergent error estimates are established. The method connects to nonconforming virtual element and pseudostress-velocity-pressure mixed formulations. Numerical experiments verify the theory.

We investigate the use of neural networks (NNs) for the estimation of hidden model parameters and uncertainty quantification from noisy observational data for inverse parameter estimation problems. We formulate the parameter estimation as a Bayesian inverse problem. We consider a parametrized system of nonlinear ordinary differential equations (ODEs), which is the FitzHugh--Nagumo model. The considered problem exhibits significant mathematical and computational challenges for classical parameter estimation methods, including strong nonlinearities, nonconvexity, and sharp gradients. We explore how NNs overcome these challenges by approximating reconstruction maps for parameter estimation from observational data. The considered data are time series of the spiking membrane potential of a biological neuron. We infer parameters controlling the dynamics of the model, noise parameters of autocorrelated additive noise, and noise modeled via stochastic differential equations, as well as the covariance matrix of the posterior distribution to expose parameter uncertainties--all with just one forward evaluation of an appropriate NN. We report results for different NN architectures and study the influence of noise on prediction accuracy. We also report timing results for training NNs on dedicated hardware. Our results demonstrate that NNs are a versatile tool to estimate parameters of the dynamical system, stochastic processes, as well as uncertainties, as they propagate through the governing ODE.

The Bauer-Furuta invariant of a family of smooth 4-manifolds is a stable cohomotopy refinement of the families Seiberg-Witten invariant and is constructed from a finite dimensional approximation of the Seiberg-Witten monopole map. We prove a general formula for the families Bauer-Furuta invariant of a fibrewise connected sum, extending Bauer's non-parameterised formula. In a subsequent paper, we will use this formula to derive a general connected sum formula for the families Seiberg-Witten invariant which incorporates both the families blow-up formula of Liu and the gluing formula of Baraglia-Konno.

Simulation-based optimization of complex systems over discrete decision spaces is a challenging computational problem. Specifically, discrete decision spaces lead to a combinatorial explosion of possible alternatives, making it computationally prohibitive to perform simulations for all possible combinations. In this work, we present a new approach to handle these issues by transforming/projecting the discrete decision space into a continuous latent space using a probabilistic model know as Variational AutoEncoders. The transformation of the decision space facilitates the implementation of Bayesian optimization (BO), which is an efficient approach that strategically navigates the space to reduce the number of expensive simulations. Here, the key observation is that points in the latent space correspond to decisions in the original mixed-discrete space, but the latent space is much easier to navigate using BO. We illustrate the benefits of our approach through a couple of case studies that aim to design complex distillation systems: the recovery of caprylic acid from water by liquid-liquid extraction and the separation of an azeotropic mixture using a thermally couple column know as extractive dividing wall column.

Quantum coloring finds applications in quantum cryptography and information. In this paper, we study the quantum chromatic numbers of Hamming graphs and a generalization of Hadamard graphs. We investigate the separation between the quantum and classical chromatic numbers of these graphs and determine the quantum chromatic numbers for some of them. For the upper bounds of the quantum chromatic numbers, we develop a linear programming approach over the Hamming scheme to construct modulus-one orthogonal representations. For the lower bounds, we determine the minimum eigenvalues for some of these graphs to derive corresponding spectral lower bounds on their quantum chromatic numbers.

Let $G$ be an additive finite abelian group, and let $\mathrm{disc}(G)$ denote the smallest positive integer $t$ with the property that every sequence $S$ over $G$ with length $|S|\geq t $ contains two nonempty zero-sum subsequences of distinct lengths. In recent years, Gao et al. established the exact value of $\mathrm{disc}(G)$ for all finite abelian groups of rank $2$ and resolved the corresponding inverse problem for the group $C_n \oplus C_n$. In this paper, we characterize the structure of sequences $S$ over $G = C_n \oplus C_{nm}$ (where $m\geq 2$) when $|S| = \mathrm{disc}(G)- 1$ and all nonempty zero-sum subsequences of $S$ have the same length.

This paper investigates the homology groups of the clique complex associated with the zero-divisor graph of a finite commutative ring. Generalizing the construction introduced by F. R. DeMeyer and L. DeMeyer, we establish a Kunneth-type formula for the homology of such complexes and provide explicit computations for products of finite local rings. As a notable application, we obtain a general method to determine the clique homology groups of Z_n and related ring products. Furthermore, we derive explicit formulas for the Betti numbers when all local factors are fields or non-fields. A complete classification of when this clique complex is Cohen-Macaulay is given, with the exception of one borderline case. Finally, our results yield a partial answer to a question posed in earlier literature, showing that certain topological spaces such as the Klein bottle and the real projective plane cannot be realized as zero-divisor complexes of finite commutative rings.

Active reconfigurable intelligent surface (RIS) emerges as an effective technique to resist the double-fading attenuation of passive RIS. By embedding with power harvesting function, it further evolves to zero-power active RIS, which can effectively enhance the flexibility of RIS deployment without external power demand. Nevertheless, existing works neglected the inherent difficulty of channel estimation (CE) for RIS-assisted systems, and the discrete phase shift constraint in practical deployment. In this paper we design a new element-wise RIS architecture and propose a distributed location-aided transmission scheme with low complexity to enhance the reflected gain for channel state information (CSI)-limited RIS-assisted near-field communications. Specifically, the new element-wise RIS provides dynamic element selection capability with low hardware resources. Based on Fresnel diffraction theory, we construct the mapping from locations in space-domain to phase distributions of waves in phase-domain and reveal the priority of elements for harvesting and reflecting. {Then, the distributed beamforming design with the phase of determine-then-align is proposed, where the estimation overhead reduction stems from exempted requirements of RIS-associated CE at base station (BS).} The asymptotic analysis indicates that the proposed scheme can achieve the optimal gain with a fixed proportion of reflective elements when RIS is large, followed by simulations to verify its superiority to other protocols.

We develop and test high-order methods for integration on surface point clouds. The task of integrating a function on a surface arises in a range of applications in engineering and the sciences, particularly those involving various integral methods for partial differential equations. Mesh-based methods require a curved mesh for high-order convergence, which can be difficult to reliably obtain on many surfaces, and most meshfree methods require the ability to integrate a set of functions (such as radial basis functions) exactly on the domain of interest; these integrals are generally not known in closed form on most surfaces. We describe two methods for integrating on arbitrary, piecewise-smooth surfaces with or without boundary. Our approaches do not require a particular arrangement of points or an initial triangulation of the surface, making them completely meshfree. We also show how the methods can be extended to handle singular integrals while maintaining high accuracy without changing the point density near singularities.

Immersive virtual reality (VR) applications impose stringent requirements on latency, energy efficiency, and computational resources, particularly in multi-user interactive scenarios. To address these challenges, we introduce the concept of spatial computing communications (SCC), a framework designed to meet the latency and energy demands of multi-user VR over distributed mobile edge computing (MEC) networks. SCC jointly represents the physical space, defined by users and base stations, and the virtual space, representing shared immersive environments, using a probabilistic model of user dynamics and resource requirements. The resource deployment task is then formulated as a multi-objective combinatorial optimization (MOCO) problem that simultaneously minimizes system latency and energy consumption across distributed MEC resources. To solve this problem, we propose MO-CMPO, a multi-objective consistency model with policy optimization that integrates supervised learning and reinforcement learning (RL) fine-tuning guided by preference weights. Leveraging a sparse graph neural network (GNN), MO-CMPO efficiently generates Pareto-optimal solutions. Simulations with real-world New Radio base station datasets demonstrate that MO-CMPO achieves superior hypervolume performance and significantly lower inference latency than baseline methods. Furthermore, the analysis reveals practical deployment patterns: latency-oriented solutions favor local MEC execution to reduce transmission delay, while energy-oriented solutions minimize redundant placements to save energy.

In this paper, we use the skein exact sequence and other techniques to compute the second-to-top term of HFK of closed 3-braids. We do it case-by-case according to Xu's classification.

In this paper we study non-standard holomorphic structures on line bundles over the quantum projective line $\mathbb{C} P^1_q$. We show that there exist infinitely many non-gauge equivalent holomorphic structures on those line bundles. This gives a negative answer to a question raised by Khalkhali, Landi, and Van Suijlekom in 2011.

To a graph $G$ one associates the binomial edge ideal $J_G$ generated by a collection of binomials corresponding to the edges of $G$. In this paper, we study the asymptotic behavior of symbolic powers of $J_G$, its lexicographic initial ideal $\mathrm{in}_<(J_G)$, and its multigraded generic initial ideal $\mathrm{gin}(J_G)$. We focus on the Waldschmidt constant, $\widehat{\alpha}$, and asymptotic regularity, $\widehat{\mathrm{reg}}$, which capture linear growth of minimal generator degrees and Castelnuovo--Mumford regularity. We explicitly compute $\widehat{\alpha}(J_G)$ and $\widehat{\alpha}(\mathrm{in}_<(J_G))$, and compare the Betti numbers of the symbolic powers of $J_G$ and $J_H$, where $H$ is a subgraph of $G$. To analyze $\mathrm{in}_<(J_G)$ and $\mathrm{gin}(J_G)$, we use the symbolic polyhedron, a convex polyhedron that encodes the elements of the symbolic powers of a monomial ideal. We determine its vertices via $G$'s induced connected subgraphs and show that $\widehat{\alpha}(\mathrm{gin}(J_G))=\widehat{\alpha}(I_G)$, where $I_G$ is the edge ideal of $G$. This yields an alternate proof of known bounds for $\widehat{\alpha}(I_G)$ in terms of $G$'s clique number and chromatic number.

This work proposes RIS-enabled channel signature modulation (RIS-CSM), a lightweight index modulation scheme for reconfigurable intelligent surfaces (RIS). An N-element RIS is partitioned into disjoint groups, each employing predetermined binary reflection patterns to generate distinct channel signatures at an $n_R$-antenna receiver, without RIS-side beamforming. Information is embedded in the indices of these signatures, enabling simple channel estimation and scalable spectral efficiency. A closed-form upper bound on error probability and capacity analysis are derived, revealing diversity order $n_R$ and coding gain proportional to N. Simulation results under Rayleigh fading validate the theoretical analysis. Moreover, simulations indicate that spatial correlation among RIS elements can improve system performance at low spectral efficiency.

Uniform asymptotic expansions are derived for the zeros of the reverse generalized Bessel polynomials of large degree $n$ and real parameter $a$. It is assumed that $-\Delta_{1} n+\frac{3}{2} \leq a \leq \Delta_{2} n$ for fixed arbitrary $\Delta_{1} \in (0,1)$ and bounded positive $\Delta_{2}$. For this parameter range at most one of the zeros is real, with the rest being complex conjugates. The new expansions are uniformly valid for all the zeros, and are shown to be highly accurate for moderate or large values of $n$. They are consequently used as initial values in a very efficient numerical algorithm designed to obtain the remaining complex zeros using Taylor series.

The Keller-Segel (KS) chemotaxis system is used to describe the overall behavior of a collection of cells under the influence of chemotaxis. However, solving the KS chemotaxis system and generating its aggregation patterns remain challenging due to the emergence of solutions exhibiting near-singular behavior, such as finite-time blow-up or concentration phenomena. Building on a Lagrangian framework of the KS system, we develop DeepLagrangian, a self-adaptive density estimation method that learns and generates aggregation patterns and near-singular solutions of the KS system in two- and three-dimensional (2D and 3D) space under different physical parameters. The main advantage of the Lagrangian framework is its inherent ability to adapt to near-singular solutions. To develop this framework, we normalize the KS solution into a probability density function (PDF), derive the corresponding normalized KS system, and utilize the property of the continuity equation to rewrite the system into a Lagrangian framework. We then define a physics-informed Lagrangian loss to enforce this framework and incorporate a flow-based generative model, called the time-dependent KRnet, to approximate the PDF by minimizing the loss. Furthermore, we integrate time-marching strategies with the time-dependent KRnet to enhance the accuracy of the PDF approximation. After obtaining the approximate PDF, we recover the original KS solution. We also prove that the Lagrangian loss effectively controls the Kullback-Leibler (KL) divergence between the approximate PDF and the exact PDF. In the numerical experiments, we demonstrate the accuracy of our DeepLagrangian method for the 2D and 3D KS chemotaxis system with/without advection.

We consider expanding systems with invariant measures that are uniformly expanding everywhere except on a small measure set and show that the limiting statistics of hitting times for zero measure sets are compound Poisson provided the limits for the cluster size distributions exist. This extends previous results from neighbourhoods around single points to neighbourhoods around zero measure sets. The assumptions require the correlations to decay at least polynomially and the non-uniformly expanding part of the iterates of the map also has to satisfy some decay condition. We also require some regularity conditions around the limiting zero measure target set.

In this paper, we show that on a compact Kähler manifold the Calabi flow can be extended as long as some space-time $L^p$ integrals of the scalar curvature are bounded.

In this paper, we study the Rasmussen-Tamagawa conjecture for abelian varieties with constrained prime power torsion. Previously, Rasmussen and Tamagawa have established the conjecture under the Generalized Riemann Hypothesis for abelian varieties of any dimension over any number field, and unconditionally for those over $\mathbb{Q}$ of dimension at most three. We prove several cases of the conjecture by giving partial refinements of their techniques. Among other things, we give an unconditional proof of the conjecture for abelian fivefolds over $\mathbb{Q}$.

We investigate explicit extreme values of the argument of the Riemann zeta-function in short intervals. As an application, we improve the result of Conrey and Turnage-Butterbaugh concerning $r$-gaps between zeros of the Riemann zeta-function.

The accurate representation of numerous physical, chemical, and biological processes relies heavily on differential equations (DEs), particularly nonlinear differential equations (NDEs). While understanding these complex systems necessitates obtaining solutions to their governing equations, the derivation of precise approximations for NDEs remains a formidable task in computational mathematics. Although established techniques such as the finite element method (FEM) have long been foundational, remarkable promise for approximating continuous functions with high efficacy has recently been demonstrated by advancements in physics-informed deep-learning feedforward neural networks. In this work, a novel application of PINNs is presented for the approximation of the challenging Electrohydrodynamic (EHD) problem. A specific $L^2$-type \textit{total loss function} is employed, notably without reliance on any prior knowledge of the exact solution. A comprehensive comparative study is conducted, juxtaposing the approximation capabilities of the proposed neural network with those of the conventional FEM. The PINN training regimen is composed of two critical steps: forward propagation for adjustments to gradient and curvature, and backpropagation for the refinement of hyperparameters. The critical challenge of identifying optimal neural network architectures and hyperparameter configurations for efficient optimization is meticulously investigated. Excellent performance is shown to be delivered by the neural network even with a limited training dataset. Simultaneously, it is demonstrated that the accuracy of the FEM can be substantially enhanced through the judicious selection of smaller mesh sizes.

We study the propagation speed of bistable traveling waves in the classical two-component diffusive Lotka-Volterra system under strong competition. From an ecological perspective, the sign of the propagation speed determines the long-term outcome of competition between two species and thus plays a central role in predicting the success or failure of invasion of an alien species into habitats occupied by a native species. Using comparison arguments, we establish sufficient conditions determining the sign of the propagation speed, which refine previously known results. In particular, we show that in the symmetric case, where the two species differ only in their diffusion rates, the faster diffuser prevails over a substantially broader parameter range than previously established. Moreover, we demonstrate that when the interspecific competition coefficients differ significantly, the outcome of competition cannot be reversed by adjusting diffusion or growth rates. These findings provide a rigorous theoretical framework for analyzing invasion dynamics, offering sharper mathematical criteria for invasion success or failure.

In [52], Parmenter and Pollicott establish an abstract criterion that gives a geometric construction of equilibrium states for a class of partially hyperbolic systems. We refine their criterion to cover a much broader class of diffeomorphisms, which include certain diffeomorphisms with exponential mixing property (with respect to volume), Katok maps and ``almost Anosov'' diffeomorphisms. As a special case, we obtain a construction of equilibrium states for ergodic partially hyperbolic affine maps/flows on homogeneous spaces, without any restrictions on the orbit growth along center directions.

Column generation and branch-and-price are leading methods for large-scale exact optimization. Column generation iterates between solving a master problem and a pricing problem. The master problem is a linear program, which can be solved using a generic solver. The pricing problem is highly dependent on the application but is usually discrete. Due to the difficulty of discrete optimization, high-performance column generation often relies on a custom pricing algorithm built specifically to exploit the problem's structure. This bespoke nature of the pricing solver prevents the reuse of components for other applications. We show that domain-independent dynamic programming, a software package for modeling and solving arbitrary dynamic programs, can be used as a generic pricing solver. We develop basic implementations of branch-and-price with pricing by domain-independent dynamic programming and show that they outperform a world-leading solver on static mixed integer programming formulations for seven problem classes.

This paper explores the asymptotic behaviour of the radii of convexity and uniform convexity for normalized Bessel functions with respect to large order. We provide detailed asymptotic expansions for these radii and establish recurrence relations for the associated coefficients. Additionally, we derive generalized bounds for the radii of convexity and uniform convexity by applying the Euler-Rayleigh inequality and potential polynomials. The asymptotic inversion method and Rayleigh sums are the main tools used in the proofs.

Renewable generators must commit to day-ahead market bids despite uncertainty in both production and real-time prices. While forecasts provide valuable guidance, rare and unpredictable extreme events (so-called black swans) can cause substantial financial losses. This paper models the nomination problem as an instance of optimal transport-based distributionally robust optimization (OT-DRO), a principled framework that balances risk and performance by accounting not only for the severity of deviations but also for their likelihood. The resulting formulation yields a tractable, data-driven strategy that remains competitive under normal conditions while providing effective protection against extreme price spikes. Using four years of Finnish wind farm and market data, we demonstrate that OT-DRO consistently outperforms forecast-based nominations and significantly mitigates losses during black swan events.

We study the Order-$k$ ($k \geq 4$) spiked tensor model for the tensor principal component analysis (PCA) problem: given $N$ i.i.d. observations of a $k$-th order tensor generated from the model $\mathbf{T} = \lambda \cdot v_*^{\otimes k} + \mathbf{E}$, where $\lambda > 0$ is the signal-to-noise ratio (SNR), $v_*$ is a unit vector, and $\mathbf{E}$ is a random noise tensor, the goal is to recover the planted vector $v_*$. We propose a normalized stochastic gradient ascent (NSGA) method with overparameterization for solving the tensor PCA problem. Without any global (or spectral) initialization step, the proposed algorithm successfully recovers the signal $v_*$ when $N\lambda^2 \geq \widetilde{\Omega}(d^{\lceil k/2 \rceil})$, thereby breaking the previous conjecture that (stochastic) gradient methods require at least $\Omega(d^{k-1})$ samples for recovery. For even $k$, the $\widetilde{\Omega}(d^{k/2})$ threshold coincides with the optimal threshold under computational constraints, attained by sum-of-squares relaxations and related algorithms. Theoretical analysis demonstrates that the overparameterized stochastic gradient method not only establishes a significant initial optimization advantage during the early learning phase but also achieves strong generalization guarantees. This work provides the first evidence that overparameterization improves statistical performance relative to exact parameterization that is solved via standard continuous optimization.

The class $A_\alpha^p$ consists of those analytic functions $f$ in the unit disc such that \[\|f\|_{\alpha,p}^p := |f(0)|^p+\int_0^1 \left(\frac{d}{dr} M_p^p(r,f)\right) (1-r^2)^{\alpha-1} \,dr < \infty,\] where $M_p^p(r,f)$ is the radial integral mean of $|f|^p$ and $0<\alpha, p <\infty$. For $\alpha>1$, $A_\alpha^p$ is the standard weighted Bergman space, and $A_1^p=H^p$. We consider $A_\alpha^p$ for $0<\alpha<1$ and show that (weighted) isometric conformal invariance extends to this range, and we also clarify the relation between $A_{\alpha}^p$ and the classical Besov spaces. Our main result is the contractive inequality $\|f\|_{\beta,q} \leq \|f\|_{\alpha,p}$, valid when $0<\alpha<\beta<\infty$ and $\alpha/p=\beta/q$. We also identify the functions for which equality is attained. We thus extend recent results of the second-named author ($1\leq \alpha<\beta$) and Llinares ($\beta=1$ and $p=2$). The extension of results from the classical range $1\leq \alpha < \infty$ to the Dirichlet range $0<\alpha <1$ uses arguments relying on analytic continuation.

Although for the most part classical, the topic of electrostatics finds to this day new applications. In this review we highlight several theoretical results on electrostatics, chosen to both illustrate general principles, and for their application in statistical mechanics and random matrix settings. The theoretical results include electrostatic potentials and energies associated with balls and hyperellipsoids in general dimension, the use of conformal mappings in two-dimensions, and the balayage measure. A number of explicit examples of their use in predicting the leading asymptotic form of certain configuration integrals and particle density in particular statistical mechanical systems are given, as well as with regards to questions relating to fluctuation formulas and (conditioned) gap probabilities.

We propose a class of numerical methods for the nonlinear Schrödinger (NLS) equation that conserves mass and energy, is of arbitrarily high-order accuracy in space and time, and requires only the solution of a scalar algebraic equation per time step. We show that some existing spatial discretizations, including the popular Fourier spectral method, are in fact energy-conserving if one considers the appropriate form of the energy density. We develop a new relaxation-type approach for conserving multiple nonlinear functionals that is more efficient and robust for the NLS equation compared to the existing multiple-relaxation approach. The accuracy and efficiency of the new schemes is demonstrated on test problems for both the focusing and defocusing NLS.

This work investigates the Sobolev regularity of solutions to perturbed fractional 1-Laplace equations. Under the assumption that weak solutions are locally bounded, we establish that the regularity properties are analogous to those observed in the superquadratic case. By introducing the threshold $\frac{p-1}{p}$, we divide the range of the parameter $s_p$ into two distinct scenarios. Specifically, for any $s_p\in \left(0, \frac{p-1}{p}\right]$ and $q\ge p$, we demonstrate that the solutions possess $W_{\rm loc}^{\gamma, q}$-regularity for all $\gamma\in \left(0, \frac{s_p p}{p-1}\right)$ and the $W_{\rm loc}^{1, q}$-regularity for any $s_p\in \left(\frac{p-1}{p}, 1\right)$ and $q\ge p$, respectively. Our analysis relies on the nonlocal finite-difference quotient method combined with a Moser-type iteration scheme, which provides a systematic approach to the regularity theory for such nonlocal and singular problems.

Let $X\subset \mathbb P^n$ be a degree $d$ hypersurface. We prove that $X$ is GIT stable if the minimal exponent $\widetilde \alpha(X)>\frac{n+1}{d}$ and GIT semistable if $\widetilde \alpha(X)=\frac{n+1}{d}$, resolving a question of Laza. Conversely, for GIT semistable cubic hypersurfaces, we prove a uniform lower bound for the minimal exponent, which implies that every such cubic has canonical singularities (and is terminal for $n\ge 6$), answering a question of Spotti-Sun. In the classical cases $(n,d)=(2,4),(2,6),(3,3),(4,3),(5,3)$, the period map from the GIT moduli is an open embedding over the stable locus with $\widetilde \alpha(X)>\frac{n+1}{d}$ and extends regularly to the Baily-Borel compactification precisely along the boundary where $\widetilde \alpha(X)=\frac{n+1}{d}$. To generalize this period map behavior in the Calabi-Yau type case $\frac{n+1}{d}=m+1\in \mathbb Z$, we introduce $m$-liminal sources and $m$-liminal centers, refining the theory of sources and log canonical centers. For an $m$-Du Bois hypersurface, we prove that the core of the limit mixed Hodge structure of any one-parameter smoothing is completely determined by the $m$-liminal source. In particular, maximal unipotent degeneration is detected by the local singularity type of the special fiber.

The aim of this work is to establish the well-posedness of fully nonlinear partial differential equations (PDE) posed on a star-shaped network, having nonlinear Kirchhoff's boundary condition at the vertex, and possibly degenerate. We obtain a comparison theorem, for discontinuous viscosity solutions, following the recent ideas obtained by Ohavi for second order problems, building test functions at the vertex solutions of Eikonal equations with well-designed coefficients. Another strong result obtained in this contribution is to show that any generalized Kirchhoff's viscosity solution introduced by Lions-Souganidis, is indeed a Kirchhoff's viscosity solution. In other terms, the values of the Hamiltonians are not required at the vertex in the analysis of these types of PDE systems.

In this paper, we obtain the optimal rigidity of dimension estimate for holomorphic functions with polynomial growth on Kähler manifolds with non-negative holomorphic bisectional curvature. There is a specific gap between the largest and the second largest dimension. We also show that the manifold attains the second largest dimension is biholomorphic to the complex Euclidean space.

This paper develops the necessary ingredients for the variational approach of initial boundary-value problems of parabolic partial differential equations on a fixed spatial domain containing evolving subdomains. In particular, we introduce function spaces for the variational solution that extend standard Sobolev-Bochner spaces to account for a coefficient associated with the time derivative that may be discontinuous across the evolving interface. We further show the density of smooth functions in these spaces by extending the mollification technique and the Reynolds transport theorem, and establish the corresponding "embedding" theory and an integration by parts formula. Finally, we prove the well-posedness of the space-time variational formulation in the natural setting using the Banach-Necas-Babuska theorem.

In this work, we present an asymptotic-preserving semi-Lagrangian discontinuous Galerkin scheme for the Boltzmann equation that effectively handles multi-scale transport phenomena. The main challenge lies in designing appropriate moments update for penalization within the semi-Lagrangian framework. Inspired by [M. Ding, J. M. Qiu, and R. Shu, Multiscale Model. Simul. 21 (2023), no. 1, 143--167], the key ingredient is utilizing the Shu-Osher form of the scheme in the implicit-explicit Runge-Kutta (IMEX-RK) setting, which enables us to capture the correct limiting system by constructing an appropriate moments update procedure. Our theoretical analysis establishes accuracy order conditions for both the IMEX-RK time integration and the new moments update step. We also employ hypocoercivity techniques to establish stability for the linearized model. Numerical experiments for various test problems validate our proposed scheme's accuracy, asymptotic-preserving property, and robustness in various regimes, which demonstrates its effectiveness for multi-scale kinetic simulations.

We provide a short proof of the $L^2$-orbital stability of a class of explicit steady Euler flows in a disk by establishing a quantitative estimate. The main idea is to exploit the conserved quantities of the Euler equation, including the kinetic energy, the enstrophy, and the moment of fluid impulse. Our result seems to suggest that more radial symmetry leads to stronger instability.

We study rational iterated preimages of the origin under unicritical maps $f_{d,c}(x)=x^d+c$. Earlier works of Faber--Hutz--Stoll and Hutz--Hyde--Krause established finiteness and conditional bounds in the quadratic case. Building on this, we prove that for $d=2$ and $c \in \mathbb Q\setminus\{0,-1\}$ there are no rational fourth preimages of the origin, and for all $d \geq 3$ there are no rational second preimages outside trivial cases. The proof relies on geometric analysis of preimage curves, the elliptic Chabauty method, and Diophantine reduction. As a result, we determine the number of rational iterated preimages of $0$ under $f_{d,c}$ for all $d\geq 2$.

Using relativized ordinal analysis, we give a proof-theoretic characterization of the provably total set-recursive-from-$\omega$ functions of KPl and related theories.

This work aims to automate the design of Multiple Gravity-Assist (MGA) transfers between planets using low-thrust propulsion. In particular, during the preliminary design phase of space missions, the combinatorial complexity of MGA sequencing is very large, and current optimization approaches require extensive experience and can take many days to simulate. Therefore, a novel optimization approach is developed here -- called the Recursive Target Body Approach (RTBA) -- that uses the hodographic-shaping low-thrust trajectory representation together with a unique combination of tree-search methods to automate the optimization of MGA sequences. The approach gradually constructs the optimal MGA sequence by recursively evaluating the optimality of subsequent gravity-assist targets. Another significant contribution to the novelty of this work is the use of parallelization in an original way involving the Generalized Island Model (GIM) that enables the use of new figures of merit to further increase the robustness and accelerate the convergence. An Earth-Jupiter transfer with a maximum of three gravity assists is considered as a reference problem. The RTBA takes 21.5 hours to find an EMJ transfer with 15.4 km/s $\Delta V$ to be the optimum. Extensive tuning improved the quality of the MGA trajectories substantially, and as a result a robust low-thrust trajectory optimization could be ensured. A distinct group of highly fit MGA sequences is consistently found that can be passed on to a higher-fidelity method. In conclusion, the RTBA can automatically and reliably be used for the preliminary optimization of low-thrust MGA trajectories.

We consider the stochastic nonlinear Schrödinger equation driven by linear multiplicative noise in the mass-supercritical case. Given arbitrary $K$ solitary waves with distinct speeds, we construct stochastic multi-solitons pathwisely in the sense of controlled rough path, which behave asymptotically as the sum of the $K$ prescribed solitons as time tends to infinity. In contrast to the mass-(sub)critical case in \cite{RSZ23}, the linearized Schrödinger operator around the ground state has more unstable directions in the supercritical case. Our pathwise construction utilizes the rescaling approach and the modulation method in \cite{CMM11}. We derive the quantitative decay rates dictated by the noise for the unstable directions, as well as the modulation parameters and remainder in the geometrical decomposition. They are important to close the key bootstrap estimates and to implement topological arguments to control the unstable directions. As a result, the temporal convergence rate of stochastic multi-solitons, which can be of either exponential or polynomial type, is related closely to the spatial decay rate of the noise and reflects the noise impact on soliton dynamics.

We investigate the asymptotic number of equivalence classes of linear codes with prescribed length and dimension. While the total number of inequivalent codes of a given length has been studied previously, the case where the dimension varies as a function of the length has not yet been considered. We derive explicit asymptotic formulas for the number of equivalence classes under three standard notions of equivalence, for a fixed alphabet size and increasing length. Our approach also yields an exact asymptotic expression for the sum of all q-binomial coefficients, which is of independent interest and answers an open question in this context. Finally, we establish a natural connection between these asymptotic quantities and certain discrete Gaussian distributions arising from Brownian motion, providing a probabilistic interpretation of our results.

A system of simple closed curves on a surface of genus $g$ is said to be sparse if their average pairwise intersection number does not exceed one. We show that the maximal size of a sparse curve systems grows roughly like a function of type $c^{\sqrt{g}}$, with $c$ between $2$ and $81938$.

The geometrical structure of PLS shrinkages is here considered. Firstly, an explicit formula for the shrinkage vector is provided. In that expression, shrinkage factors are expressed a averages of a set of basic shrinkages that depend only on the data matrix. On the other hand, the weights of that average are multilinear functions of the observed responses. That representation allows to characterise the set of possible shrinkages and identify extreme situations where the PLS estimator has an highly nonlinear behaviour. In these situations, recently proposed measures for the degrees of freedom (DoF), that directly depend on the shrinkages, fail to provide reasonable values. It is also shown that the longstanding conjecture that the DoFs of PLS always exceeds the number PLS directions does not hold.

We investigate a class of elliptic and parabolic partial differential equations driven by p(u) laplacian. This dependence necessitates the use of variable exponent Sobolev spaces specifically tailored to the anisotropic framework. For the elliptic case, we establish the existence of a weak solution by employing the theory of pseudomonotone operators in conjunction with suitable approximation techniques. In the parabolic setting, the existence of a weak solution is obtained via a time discretization scheme and Schauder fixed-point theorem, supported by a priori estimates and compactness arguments.

Let $R$ be a discrete valuation ring, with valuation $v \colon R \twoheadrightarrow \mathbb{Z}_{\ge 0} \cup \{\infty\}$ and residue field $k$. Let $H$ be a hypersurface $\operatorname{Proj}(R[x_0,\ldots,x_n]/\langle f \rangle)$. Let $H_k$ be the special fiber, and let $(H_k)_{\mathrm{sing}}$ be its singular subscheme. Let $\Delta(f)$ be the discriminant of $f$. We use Zariski's main theorem and degeneration arguments to prove that $v(\Delta(f))=1$ if and only if $H$ is regular and $(H_k)_{\mathrm{sing}}$ consists of a nondegenerate double point over $k$. We also give lower bounds on $v(\Delta(f))$ when $H_k$ has multiple singularities or a positive-dimensional singularity.

This article discusses the convergence properties of the Max Product and Max Min variants of Durrmeyer type exponential sampling series. We first establish pointwise and uniform convergence of both operators in the space of log uniformly continuous and bounded functions. The rates of convergence are then analyzed in terms of the logarithmic modulus of continuity. Additionally, the approximation errors of the proposed operators are examined using a variety of kernel functions. Finally, graphical illustrations are provided to demonstrate the convergence behavior of both operators.

Time series aggregation (TSA) methods aim to construct temporally aggregated optimization models that accurately represent the output space of their full-scale counterparts while using a significantly reduced dimensionality in the input space. This paper presents the first approach that achieves an exact TSA of a full-scale power system model -- even in the presence of energy storage time-coupling constraints -- by leveraging active constraint sets and dual information. This advances the state of the art beyond existing TSA approaches, which typically cannot guarantee solution accuracy or rely on iterative procedures to determine the required number of representative periods. To bridge the gap between our theoretical analysis and their practical application, we employ machine learning approaches, i.e., classification and clustering, to inform TSA in models that co-schedule variable renewable energy sources and energy storage. Numerical results demonstrate substantially improved computational performance relative to the full-scale model, while maintaining high solution accuracy.

We introduce new classes of general monotone sequences and study their properties. For functions whose Fourier coefficients belong to these classes, we establish Hardy-Littlewood-type theorems.

We consider Schr{ö}dinger equations with logarithmic nonlinearity and bilinear controls, posed on $\mathbb{T}^d$ or $\mathbb{R}^d$. We prove their small-time global $L^2$-approximate controllability. The proof consists in extending to this nonlinear framework the approach introduced by the first and third authors in \cite{beauchard-pozzoli2} to control the linear equation: it combines the small-time controllability of phases and gradient flows. Due to the nonlinearity, the required estimates are more difficult to establish than in the linear case. The proof here is inspired by WKB analysis. This is the first result of (small-time) global approximate controllability, for nonlinear Schr{ö}dinger equations, with bilinear controls.

The use of the Preconditioned Conjugate Gradient (PCG) method for computing the Generalized Least Squares (GLS) estimator of the General Linear Model (GLM) is considered. The GLS estimator is expressed in terms of the solution of an augmented system. That system is solved by means of the PCG method using an indefinite preconditioner. The resulting method iterates a sequence Ordinary Least Squares (OLS) estimations that converges, in exact precision, to the GLS estimator within a finite number of steps. The numerical and statistical properties of the estimator computed at an intermediate step are analytically and numerically studied. This approach allows to combine direct methods, used in the OLS step, with those of iterative methods. This advantage is exploited to design PCG methods for the estimation of Constrained GLMs and of some structured multivariate GLMs. The structure of the matrices involved are exploited as much as possible, in the OLS step. The iterative method then solves for the unexploited structure. Numerical experiments shows that the proposed methods can achieve, for these structured problems, the same precision of state of the art direct methods, but in a fraction of the time.

Let $L/K$ be any finite separable extension with normal closure $\widetilde{L}/K$. An extension $L'/K$ is said to be $\textit{parallel to $L/K$}$ if $L'$ is an intermediate field of $\widetilde{L}/K$ with $[L':K]=[L:K]$. We study the following question -- Given that $L/K$ admits a Hopf--Galois structure of type $N$, does it imply that every extension parallel to $L/K$ also admits a Hopf--Galois structure of type $N$? We completely solve this problem when the degree $[L:K]$ is a prime power and the type $N$ is cyclic. Our approach is group-theoretic and uses the work of Greither--Pareigis and Byott.

In this paper we discuss a new method to blend fractal attractors using the code map for the IFS formed by the Hutchinson--Barnsley operators of a finite family of hyperbolic IFSs. We introduce a parameter called blending coefficient to measure the similarity between the blended set and each one of the original attractors. We also introduce a discrete approximation algorithm and prove a rigorous error estimation used to approximate these new objects. Several simulation results are provided illustrating our techniques.

For $s\in(0,1)$ and an open bounded set $\Omega\subset\mathbb R^n$, we prove existence and uniqueness of absolute minimisers of the supremal functional $$E_\infty(u)=\|(-\Delta)^s u\|_{L^\infty(\mathbb R^n)},$$ where $(-\Delta)^s$ is the Fractional Laplacian of order $s$ and $u$ has prescribed Dirichlet data in the complement of $\Omega$. We further show that the minimiser $u_\infty$ satisfies the (fractional) PDE $$ (-\Delta)^s u_\infty=E_\infty(u_\infty)\,\mathrm{sgn}f_\infty \qquad\mbox{in }\Omega, $$ for some analytic function $f_\infty\in L^1(\Omega)$ obtained as the restriction of an $s$-harmonic measure $\mu$ in $\Omega$.

We investigate endoartinian modules, which satisfy the descending chain condition on endoimages, and establish new characterizations that unify classical and generalized chain conditions. Over commutative rings, endoartinianity coincides with rings satisfying the strongly ACCR* with dim(R) = 0 and strongly DCCR* conditions. For principally injective rings, the endoartinian and endonoetherian rings are equivalent. Addressing a question of Facchini and Nazemian, we provide a condition under which isoartinian and Noetherian rings coincide, and we classify semiprime endoartinian rings as finite products of matrix rings over a division ring. We further show that endoartinianity is equivalent to the Kothe rings over principal ideal rings with central idempotents, and characterize such rings as finite products of artinian uniserial rings.

We consider linear model reduction in both the control and state variables for unconstrained linear-quadratic optimal control problems subject to time-varying parabolic PDEs. The first-order optimality condition for a state-space reduced model naturally leads to a reduced structure of the optimal control. Thus, we consider a control- and state-reduced problem that admits the same minimizer as the solely state-reduced problem. Lower and upper \emph{a posteriori} error bounds for the optimal control and a representation for the error in the optimal function value are provided. These bounds are used in an adaptive algorithm to solve the control problem. We prove its convergence and numerically demonstrate the advantage of combined control and state space reduction.

We consider a new method for estimating the parameters of univariate Gaussian mixture models. The method relies on a nonparametric density estimator $\hat{f}_n$ (typically a kernel estimator). For every set of Gaussian mixture components, $\hat{f}_n$ is used to find the best set of mixture weights. That set is obtained by minimizing the $L_2$ distance between $\hat{f}_n$ and the Gaussian mixture density with the given component parameters. The densities together with the obtained weights are then plugged in to the likelihood function, resulting in the so-called pseudo-likelihood function. The final parameter estimators are the parameter values that maximize the pseudo-likelihood function together with the corresponding weights. The advantages of the pseudo-likelihood over the full likelihood are: 1) its arguments are the means and variances only, mixture weights are also functions of the means and variances; 2) unlike the likelihood function, it is always bounded above. Thus, the maximizer of the pseudo-likelihood function -- referred to as the pseudo-likelihood estimator -- always exists. In this article, we prove that the pseudo-likelihood estimator is strongly consistent.

We construct the Bruhat-Tits stratification of the ramified unitary splitting Rapoport-Zink space, with the level being the stabilizer of a vertex lattice. To determine certain local properties of the Bruhat-Tits strata, we develop a theory of the strata splitting models. To study their global structure, we establish an explicit isomorphism between the Bruhat-Tits strata and certain (modified) Deligne-Lusztig varieties.

Given two distinct complex Hadamard matrices belonging to the same equivalence class generated by the tensor products of Fourier matrices, we show that if the corresponding Hadamard subfactors are conjugate, then their intersection is a factor with finite Jones index. We compute the index of the intersection explicitly and determine its relative commutant. Furthermore, we precisely characterize when these intersections give rise to vertex model subfactors, thereby extending our earlier results in low dimensions. As an application, we derive an explicit formula for the Connes-Størmer relative entropy associated with these intersections. These results reveal how the internal algebraic structure of complex Hadamard matrices governs the relative position and entropic behaviour of the subfactors.

This work studies circle-geometry methods through their application to a main theorem about circles tangent twice to a conic. The authors investigate the Sharygin point -- a point lying in the pencil of two non-intersecting circles -- and explore its properties. These properties are applied to solve several olympiad problems, such as problems from MGO 2024 and the Croatian IMO selection. The paper also presents a simplified version of the main theorem and gives two different proofs: one using Sharygin points and another using Lobachevsky (hyperbolic) geometry. The article explores relation between Lorenz transformations of Minkowski space-time and a certain transformation of circles in hyperbolic geometry. The paper demonstrates the effectiveness of combining classical planimetry with ideas of non-Euclidean geometry for solving difficult problems involving circle tangencies.

We extend flow matching to ensembles of linear systems in both deterministic and stochastic settings. Averaging over system parameters induces memory leading to a non-Markovian interpolation problem for the stochastic case. In this setting, a control law that achieves the distributional controllability is characterized as the conditional expectation of a Volterra-type control. This conditional expectation in the stochastic settings motivates an open-loop characterization in the noiseless-deterministic setting. Explicit forms of the conditional expectations are derived for special cases of the given distributions and a practical numerical procedure is presented to approximate the control inputs. A by-product of our analysis is a numerical split between the two regimes. For the stochastic case, history dependence is essential and we implement the conditional expectation with a recurrent network trained using independent sampling. For the deterministic case, the flow is memoryless and a feedforward network learns a time-varying gain that transports the ensemble. We show that to realize the full target distribution in this deterministic setting, one must first establish a deterministic sample pairing (e.g., optimal-transport or Schrodinger-bridge coupling), after which learning reduces to a low-dimensional regression in time.

Many logical properties are known to be undecidable for normal modal logics, with few exceptions such as consistency and coincidence with $\mathsf{K}$. This paper shows that the property of being a union-splitting in $\mathsf{NExt}\mathsf{K}$, the lattice of normal modal logics, is decidable, thus answering the open problem [WZ07, Problem 2]. This is done by providing a semantic characterization of union-splittings in terms of finite modal algebras. Moreover, by clarifying the connection to union-splittings, we show that in $\mathsf{NExt}\mathsf{K}$, having a decidable axiomatization problem and being a (un)decidable formula are also decidable. The latter answers [CZ97, Problem 17.3] for $\mathsf{NExt}\mathsf{K}$.

This work is on surfaces with a constant ratio of principal curvatures. These CRPC surfaces generalize minimal surfaces but are much more challenging to construct. We propose a construction of a family of such surfaces containing a given minimal surface without flat points. This leads to a partial solution of Plateau's problem for CRPC surfaces. We obtain analogous results in isotropic geometry. This work illustrates a general approach to solving Euclidean problems by starting with their isotropic analogs. Besides, we apply the method of successive approximations and analytic majorization.

We offer a new proof (and review some known proofs) of Cantor's Powerset Theorem (1891), which concerns the non-existence of a surjective function from a set onto its powerset.

We construct algebraic surfaces with a large number of type A singularities. Bivariate polynomials presented in previous works for the construction of nodal surfaces and certain families of Belyi polynomials are used. In some cases explicit expressions in terms of classical Jacobi polynomials are obtained.

We briefly explain how to implement the morphisms in our paper ``Natural representations of black box groups encrypting $SL_2(\mathbb{F})$" and provide some examples.

Let $G$ be a finitely generated malabelian group, let $A\leq\mathrm{Out}(G)$ be a finitely generated subgroup, and let $\Gamma_{G,A}$ denote the preimage of $A$ in $\mathrm{Aut}(G)$. We give a general criterion for the linearity of $\Gamma_{G,A}$ in terms of surjections from $G$ to finite simple groups of Lie type.

Conventional beamforming with fixed-orientation antenna (FOA) arrays may struggle to effectively enhance signal and/or suppress interference due to significant variations in antenna directive gains over different steering angles. To break this limitation, we investigate in this paper the rotatable antenna (RA)-enhanced single/multi-beam forming by exploiting the new spatial degrees of freedom (DoFs) via antennas' rotation optimization. Specifically, the antenna rotation vector (ARV) and antenna weight vector (AWV) are jointly optimized to maximize the minimum array gain over signal directions, subject to a given constraint on the maximum array gain over interference directions. For the special case of single-beam forming without interference, the optimal ARV is derived in closed-form with the maximum ratio combining (MRC) beamformer applied to the AWV. For the general case of multi-beam forming, we propose an efficient alternating optimization (AO) algorithm to find a high-quality suboptimal solution by iteratively optimizing one of the ARV and AWV with the other being fixed. Simulation results demonstrate that the proposed RA-based scheme can significantly outperform the traditional FOA-based and isotropic antenna (IA)-based schemes in terms of array gain.

We define and study certain linear orders on chainable continua. Those orders depend on a sequence of chains obtained from definition of chainability and on a fixed non-principal ultrafilter on the set of natural numbers. An alternative method of defining linear orders on a chainable continuum $X$ uses representation of $X$ as an inverse sequence of arcs and fixed non-principal ultrafilter on $\mathbb{N}$. We compare those two approaches. We prove that there exist exactly $2$ distinct ultrafilter orders on any arc, exactly $4$ distinct ultrafilter orders on the Warsaw sine curve, and exactly $2^{\mathfrak{c}}$ distinct ultrafilter orders on the Knaster continuum. We study the order type of various chainable continua equipped with an ultrafilter order and prove that a chainable continuum $X$ is Suslinian if and only if for every ultrafilter order $\leq_{\mathcal{U}}^{\mathcal{D}}$ on $X$ the space $X$ with an order topology, generated by the order $\leq_{\mathcal{U}}^{\mathcal{D}}$, is ccc. We study also descriptive complexity of ultrafilter orders on chainable continua. We prove that the existence of closed ultrafilter order characterizes the arc and we show that for Suslinian chainable continua, any ultrafilter order is both of type $F_{\sigma}$ and $G_{\delta}$. On the other hand, we prove that there is no analytic and no co-analytic ultrafilter order on the Knaster continuum.

The distinct dot products problem, a variation on the Erdős distinct distance problem, asks "Given a set $P_n$ of $n$ points in $\mathbb{R}^2$, what is the minimum number $|D(P_n)|$ of distinct dot products formed between them, asymptotically?" The best proven lower-bound is $|D(P_n)| \gtrsim n^{2/3+7/1425}$, due to work by Hanson$\unicode{x2013}$Roche-Newton$\unicode{x2013}$Senger, and a recent improvement by Kokkinos. However, the slowest-scaling known constructions have $|D(P_n)|\sim n$, leaving quite a large gap in the bound. Finding a sublinearly-scaling construction, or disproving its existence, would narrow this gap. We provide a condition that a sequence of point configurations $(P_n)_{n \in \mathbb{N}}$ must satisfy in order for $|D(P_n)|$ to scale 'slowly' i.e. $|D(P_n)| \ll n^{3/4}$. Namely, we prove that any such configuration must contain a point-rich line that gets arbitrarily 'dense' as the sequence progresses.

We study the time-harmonic Maxwell equations on bounded Lipschitz domains with an impedance boundary condition. The impedance coefficient can be matrix valued such that, in particular, a polarization dependent impedance is modeled. We derive a Fredholm alternative for this system. As a consequence, we obtain the existence of weak solutions for arbitrary sources when the frequency is not a resonance frequency. Our analysis covers the case of singular impedance coefficients.

A map between manifolds induces stratifications of both the source and the target according to the occurring multisingularities. In this paper, we study universal expressions-called higher Thom polynomials-that describe the Segre-Schwartz-MacPherson class of such multisingularity loci. We prove a Structure Theorem reducing these Thom polynomials to the data of a linear series associated with each multisingularity. The series corresponding to the empty multisingularity, referred to as the Master Series, plays a distinguished role. Motivated by connections with geometric representation theory, we further prove an Interpolation Theorem that allows Thom polynomials to be computed algorithmically within Mather's range of nice dimensions. As an application, we derive an explicit formula for the image Milnor number of quasihomogeneous germs, providing one side of the celebrated Mond conjecture, computable up to the theoretical bound.

In this paper, we prove that for any given closed contact manifold, there exists an infinite-dimensional space of Riemannian metrics which can be identified with the space of bundle metrics on the induced contact distribution. For each such metric, and for all energy levels, the number of embedded periodic orbits of the corresponding magnetic geodesic flow grows at least as fast as the number of geometrically distinct periodic Reeb orbits of period less than $t$. As a corollary, we deduce that for every closed 3-manifold which is not a graph manifold, there exists an open $C^1$-neighborhood of the set of nondegenerate contact forms such that for each contact form in this neighborhood, there exists an infinite-dimensional space of Riemannian metrics as above. For the corresponding magnetic systems, the number of prime closed magnetic geodesics grows at least exponentially on all energy levels. Consequently, the restriction of the magnetic geodesic flow to any energy surface has positive topological entropy.

Dekimpe and Ongenae constructed infinitely many pairwise non-isomorphic complete left-symmetric structures on $\mathbb{R}^n$ for $n\geq 6$. In this paper, we construct a family of complete left-symmetric structures on the cotangent Lie algebra $T^*\mathfrak{g}$ of a certain $n$-dimensional almost abelian nilpotent Lie algebra $\mathfrak{g}$ and give a condition under which two left-symmetric structures in this family are isomorphic. As a consequence of this result, we obtain infinitely many pairwise non-isomorphic left-symmetric structures on $T^{*}\mathfrak{g}$. As an application of this construction, we also obtain infinitely many symplectic structures on $T^{*}\mathfrak{g}$ which are pairwise non-symplectomorphic up to homothety.

In this article we prove global propagation of analyticity in finite time for solutions of semilinear Schrödinger equations with analytic nonlinearity from a region $\omega$ where the Geometric Control Condition holds. Our approach refines a recent technique introduced by Laurent and the author, which combines control theory techniques and Galerkin approximation, to propagate analyticity in time from a zone where observability holds. As a main consequence, we obtain unique continuation for subcritical semilinear Schrödinger equations on compact manifolds of dimension $2$ and $3$ when the solution is assumed to vanish on $\omega$. Furthermore, semiglobal control and stabilization follow only under the Geometric Control Condition on the observation zone. In particular, this answers in the affirmative an open question of Dehman, Gérard, and Lebeau from $2006$ for the nonlinear case.

We introduce a polyanalytic extension of the Gaussian radial basis function (RBF) kernel by computing the action of the convolution operator on normalized Hermite functions. In particular, using the Zaremba-Bergman formula we derive an explicit closed form for this new reproducing kernel function. We then establish an isomorphism relating the reproducing kernel Hilbert space induced by the polyanalytic Gaussian RBF kernel with the corresponding polyanalytic Fock space. Moreover, we provide a characterization of polyanalytic Gaussian RBF spaces in terms of a Landau-type operator. In addition, we investigate the polyanalytic counterpart of the Weyl operator, which leads to applications involving the Christoffel-Darboux formula for Hermite polynomials and Mehler's kernel. Finally, we discuss the analogue of the Weyl operator in the context of the polyanalytic Gaussian RBF setting.

We prove that for every $t \in \mathbb{N}$, the graph $K_{2,t}$ satisfies the fat minor conjecture of Georgakopoulos and Papasoglu: for every $K\in \mathbb{N}$ there exist $M,A\in \mathbb{N}$ such that every graph with no $K$-fat $K_{2,t}$ minor is $(M,A)$-quasi-isometric to a graph with no $K_{2,t}$ minor. We use this to obtain an efficient algorithm for approximating the minimal multiplicative distortion of any embedding of a finite graph into a $K_{2,t}$-minor-free graph, answering a question of Chepoi, Dragan, Newman, Rabinovich, and Vaxès from 2012.

Magnetic Resonance Imaging (MRI) is essential for noninvasive generation of high-quality images of human tissues. Accurate segmentation of MRI data is critical for medical applications like brain anatomy analysis and disease detection. However, challenges such as intensity inhomogeneity, noise, and artifacts complicate this process. To address these issues, we propose a three-step framework exploiting the idea of Cartoon-Texture evolution to produce a denoised and debiased MR image. The first step involves identifying statistical information about the nature of the noise using a suitable image decomposition. In the second step, a multiplicative intrinsic component model is applied to a smother version of the image, simultaneously reconstructing the bias and removing noise using noise information from the previous step. At the final step, standard clustering techniques are used to create an accurate segmentation. Additionally, we present a convergence analysis of the ADMM scheme for solving the nonlinear optimization problem with multiaffine constraints resulting from the second step. Numerical tests demonstrate the effectiveness of our framework, especially in noisy brain segmentation, both from a qualitative and a quantitative viewpoint, compared to similar methods.

In this paper, we investigate channel estimation for reconfigurable intelligent surface (RIS) empowered millimeter-wave (mmWave) multi-user single-input multiple-output communication systems using low-resolution quantization. Due to the high cost and power consumption of analog-to-digital converters (ADCs) in large antenna arrays and for wide signal bandwidths, designing mmWave systems with low-resolution ADCs is beneficial. To tackle this issue, we propose a channel estimation design using task-based quantization that considers the underlying hybrid analog and digital architecture in order to improve the system performance under finite bit-resolution constraints. Our goal is to accomplish a channel estimation task that minimizes the mean squared error distortion between the true and estimated channel. We develop two types of channel estimators: a cascaded channel estimator for an RIS with purely passive elements, and an estimator for the separate RIS-related channels that leverages additional information from a few semi-passive elements at the RIS capable of processing the received signals with radio frequency chains. Numerical results demonstrate that the proposed channel estimation designs exploiting task-based quantization outperform purely digital methods and can effectively approach the performance of a system with unlimited resolution ADCs. Furthermore, the proposed channel estimators are shown to be superior to baselines with small training overhead.

Via a new isoparametric foliation in $\mathbb{S}^n(1) \times \mathbb{S}^n(1)$, we find many new area-minimizing cones by applying Lawlor's curvature criterion, especially codimension-two area-minimizing cones in $\mathbb{R}^{2n+2}$ for $n\geq 63$. This new isoparametric foliation is a further restriction of the Ferus-Karcher-Münzner isoparametric foliation to $\mathbb{S}^n(1) \times \mathbb{S}^n(1) \subset \mathbb{S}^{2n+1}(\sqrt{2})$, which can also be defined on general $\mathbb{S}^n(a) \times \mathbb{S}^n(b)(a>0,b>0)$, and it extends the recent classification results of F. Urbano for $n=2$.

We study rank 2 torus-equivariant torsion-free sheaves on the complex projective space. For reflexive sheaves we derive a simple formula for the Chern polynomial, and in the general torsion-free case we introduce an iterative construction method based on elementary injections, allowing us to prescribe Chern classes. This yields infinite families of explicit examples on $\mathbb{P}^4$ and $\mathbb{P}^5$, and establishes existence on $\mathbb{P}^n$ for all $n\geq 3$, with Chern classes satisfying all known constraints arising from locally freeness and indecomposability. We also provide simple obstructions for smoothability.

A partial field is an algebraic object that allows one to simultaneously abstract several different representability properties of matroids. In this paper we study partial fields as algebraic objects in their own right. We characterize the weak and strong characteristic sets of partial fields and show that the class of partial fields is not well-quasi ordered. We provide a new proof that the lift operator of a partial field is idempotent. We also provide a relation between the fundamental elements of a partial field and its Dowling lift, and show that the Dowling lift operator is idempotent.

We consider a pure jump process $\{X_t\}_{t\ge 0}$ with values in a finite state space $S= \{1, \ldots, d\}$ for which the jump rates at time instant $t$ depend on the occupation measure $L_t \doteq t^{-1} \int_0^t \delta_{X_s}\,ds$. Such self-interacting chains arise in many contexts within statistical physics and applied probability. Under appropriate conditions, a large deviation principle is established for the pair $(L_t, R_t)$, as $t \to \infty$, where $R_t$ is the empirical flux process associated with the jump process. We show that the rate function takes a simple form that can be viewed as a dynamical generalization of the classical Donsker and Varadhan rate function for the analogous quantities in the setting of Markov processes, in particular, unlike the Markovian case, the rate function is not convex. Since the state process is non-Markovian, different techniques are needed than in the setting of Donsker and Varadhan and our proofs rely on variational representations for functionals of Poisson random measures and stochastic control methods.

Modeling heterogeneous and multi-lane traffic flow is essential for understanding and controlling complex transportation systems. In this work, we consider three vehicle populations: two classes of human-driven vehicles (cars and trucks) and autonomous vehicles, the latter characterized by controlled acceleration. Compared to single-population models, multi-population modeling poses greater challenges, primarily due to the increased number of parameters required to describe lane-changing behavior and the added complexity in passing to the mean-field limit. We model multi-lane traffic as a hybrid dynamical system, combining continuous dynamics within each lane and discrete events corresponding to lane-changing maneuvers. We then formulate and analyze the optimal control problem associated with such hybrid systems from both microscopic and mesoscopic perspectives. Using techniques from $\Gamma$-convergence, we prove the existence of solutions to the optimal control problem in the mean-field limit of a finite-dimensional hybrid system. Finally, we present numerical simulations illustrating the impact of trucks on overall traffic efficiency.

This study presents a high-order, space-time coupled arbitrary Lagrangian Eulerian (ALE) compact gas-kinetic scheme (GKS) for the shallow water equations on moving unstructured meshes. The proposed method preserves both the geometric conservation law (GCL) and the well-balanced property. Mesh motion effects are directly incorporated by formulating numerical fluxes that account for the spatial temporal nonuniformity of the flow field and the swept area of moving cell interfaces. This allows temporal updates to be performed on the physical moving mesh, avoiding data remapping. The compact GKS provides time accurate evolution of flow variables and fluxes, enabling the scheme to achieve second-order temporal accuracy within a single stage. To consistently treat bottom topography on moving meshes, an evolution equation for the topography is established and discretized using a compatible space-time scheme, in which the fluxes induced by mesh motion are computed accurately. Mathematical proofs demonstrating the GCL preserving and well-balanced properties of the proposed ALE formulation are also provided. For improved accuracy and robustness, a nonlinear fourth-order compact reconstruction technique is employed. A comprehensive set of numerical experiments verifies the scheme's theoretical properties and demonstrates its accuracy, stability, and effectiveness in simulating complex shallow-water flow problems.

The main result of this article is a geometric interpretation of magnitude, a real-valued invariant of metric spaces. We introduce a Euclidean embedding of a (suitable) finite metric space $X$ such that the magnitude of $X$ can be expressed in terms of the `circumradius' of its embedding $S$. The circumradius is the smallest $r$ for which the $r$-thickening of $S$ is contractible. We give three applications: First, we describe the asymptotic behaviour of the magnitude of $tX$ as $t\rightarrow \infty$, in terms of the circumradius. Second, we develop a matrix theory for magnitude that leads to explicit relations between the magnitude of $X$ and the magnitude of its subspaces. Third, we identify a new regime in the limiting behaviour of $tX$, and use this to show submodularity-type results for magnitude as a function on subspaces.

Let $(X,\omega)$ be a compact Hermitian manifold of dimension $n$. We derive an $L^\infty$-estimate for bounded solutions to the complex $m$-th Hessian equations on $X$, assuming a positive right-hand side in the Orlicz space $L^{\frac{n}{m}}(\log L)^n(h\circ\log \circ \log L)^n$, where the associated weight satisfies Kołodziej's Condition. Building upon this estimate, we then establish the existence of continuous solutions to the complex Hessian equation under the prescribed assumptions.

Datasets often possess an intrinsic multiscale structure with meaningful descriptions at different levels of coarseness. Such datasets are naturally described as multi-resolution clusterings, i.e., not necessarily hierarchical sequences of partitions across scales. To analyse and compare such sequences, we use tools from topological data analysis and define the Multiscale Clustering Bifiltration (MCbiF), a 2-parameter filtration of abstract simplicial complexes that encodes cluster intersection patterns across scales. The MCbiF can be interpreted as a higher-order extension of Sankey diagrams and reduces to a dendrogram for hierarchical sequences. We show that the multiparameter persistent homology (MPH) of the MCbiF yields a finitely presented and block decomposable module, and its stable Hilbert functions characterise the topological autocorrelation of the sequence of partitions. In particular, at dimension zero, the MPH captures violations of the refinement order of partitions, whereas at dimension one, the MPH captures higher-order inconsistencies between clusters across scales. We demonstrate through experiments the use of MCbiF Hilbert functions as topological feature maps for downstream machine learning tasks. MCbiF feature maps outperform information-based baseline features on both regression and classification tasks on synthetic sets of non-hierarchical sequences of partitions. We also show an application of MCbiF to real-world data to measure non-hierarchies in wild mice social grouping patterns across time.

Increasingly in recent years, probabilistic computation has been investigated through the lenses of categorical algebra, especially via string diagrammatic calculi. Whereas categories of discrete and Gaussian probabilistic processes have been thoroughly studied, with various axiomatisation results, more expressive classes of continuous probability are less understood, because of the intrinsic difficulty of describing infinite behaviour by algebraic means. In this work, we establish a universal construction that adjoins infinite tensor products, allowing continuous probability to be investigated from discrete settings. Our main result applies this construction to $\mathsf{FinStoch}$, the category of finite sets and stochastic matrices, obtaining a category of locally constant Markov kernels, where the objects are finite sets plus the Cantor space $2^{\mathbb{N}}$. Any probability measure on the reals can be reasoned about in this category. Furthermore, we show how to lift axiomatisation results through the infinite tensor product construction. This way we obtain an axiomatic presentation of continuous probability over countable powers of $2=\lbrace 0,1\rbrace$.

This study examines a fully parabolic predator-prey chemo-alarm-taxis system under homogeneous Neumann boundary conditions in a bounded domain $\Omega \subset \mathbb{R}^n$ with a smooth boundary $\partial\Omega$. Under specific parameter conditions, it is shown that the system admits a unique, globally bounded classical solution. The convergence of the solution is established through the construction of an appropriate Lyapunov functional. In addition, numerical simulations are presented to validate the asymptotic behaviour of the solution. The results highlight the significant role of chemotaxis and alarm-taxis coefficients in determining the existence and stability of predator-prey models, as discussed in the literature.

This article introduces the Generalized Fourier Series (GFS), a novel spectral method that extends the clas- sical Fourier series to non-periodic functions. GFS addresses key challenges such as the Gibbs phenomenon and poor convergence in non-periodic settings by decomposing functions into periodic and aperiodic com- ponents. The periodic part is represented using standard Fourier modes and efficiently computed via the Fast Fourier Transform (FFT). The aperiodic component employs adaptive, low-rank sinusoidal functions with non-harmonic modes, dynamically tuned to capture discontinuities and derivative jumps across domain boundaries. Unlike conventional Fourier extension methods, GFS achieves high accuracy without requiring compu- tational domain extensions, offering a compact and efficient representation of non-periodic functions. The adaptive low-rank approach ensures accuracy while minimizing computational overhead, typically involving additional complex modes for the aperiodic part. Furthermore, GFS demonstrates a high-resolution power, with degrees of freedom comparable to FFT in periodic domains, and maintains N log2(N) computational complexity. The effectiveness of GFS is validated through numerical experiments, showcasing its ability to approximate functions and their derivatives in non-periodic domains accurately. With its robust framework and minimal computational cost, GFS holds significant potential for advancing applications in numerical PDEs, signal processing, machine learning, and computational physics by providing a robust and efficient tool for high-accuracy function approximations.

We establish sharp asymptotic bounds for the critical intensity of the Finitary Random Interlacements (FRI) model in four and higher dimensions with general trajectory length distributions. Our proof reveals that the construction of near-critical FRI clusters in four and higher dimensions is essentially analogous to a Galton-Watson process, whose expected number of offspring corresponds to the capacity of a random walk killed at the given length.

Let $\PSp(n,1)$ denote the isometry group of quaternionic hyperbolic space $\h^n$. A pair of elements $(g_1,g_2)$ in $\PSp(n,1)$ is said to be \emph{strongly doubly reversible} if $(g_1,g_2)$ and $(g_1^{-1},g_2^{-1})$ belong to the same simultaneous conjugation orbit of $\PSp(n,1)$, and a conjugating element can be chosen to have order two. Equivalently, there exist involutions $i_1,i_2,i_3 \in \PSp(n,1)$ such that $g_1 = i_1 i_2,~ g_2 = i_1 i_3$. We prove that the set of such pairs has Haar measure zero in $\PSp(n,1) \times \PSp(n,1)$. The same result also holds for $\PSp(n) \times \PSp(n)$ for $n\geq 2$. In the special case $n=1$, we show that every pair of elements in $\PSp(1)$ is strongly doubly reversible. Using elementary quaternionic analysis for $\Sp(1)$, we also provide a very short proof of a theorem of Basmajian and Maskit, in Trans. Amer. Math. Soc. 364 (2012), no. 9, 5015--5033, which states that every pair of elements in ${\rm SO}(4)$ is strongly doubly reversible. Furthermore, we derive necessary conditions under which a pair of hyperbolic elements is strongly doubly reversible in $\PSp(1,1)$.

In this paper, we study delay differential equations involving the Schwarzian derivative $S(f,z)$, expressed in the form \begin{equation*} f(z+1)f(z-1) + a(z)S(f,z) =R(z,f(z))= \frac{P(z,f(z))}{Q(z,f(z))} \end{equation*} where $a(z)$ is rational, $P(z,f)$ and $Q(z,f)$ are coprime polynomials in $f$ with rational coefficients. Our main result shows that if a subnormal transcendental meromorphic solution exists, then the rational function $R(z,f)=P(z,f)/Q(z,f)$ satisfies $°_fR\leq 7$ and $°_fP\leq °_fQ +2$, where $°_fR =\max\{°_fP, °_fQ\}.$ Furthermore, for any rational root $b_1$ of $Q(z,f)$ in $f$ with multiplicity $k$, we show that $k \leq 2$. Finally, a classification of such equations is provided according to the multiplicity structure of the roots of $Q(z,f)$. Some examples are given to support these results.

In this article we consider group actions on compact Riemann surfaces and their topological classification. We address this problem for pairs $(S, N)$ where $S$ is a compact Riemann surface endowed with a group of automorphisms $N \cong \mathbb{Z}_k^m$ such $S/N$ has signature $(0;k,\stackrel{n+1}{\ldots},k)$, where $n, k \geqslant 2$ and $1 \leqslant m \leqslant n$ are integers. We further assume the existence of extra automorphisms, namely, a group $G$ with $N \lhd G \leqslant \mathrm{Aut}(S)$ and analyze the induced permutational action of $G/N$ on the cone points of $S/N$. To describe such actions up to topological equivalence, we employ the generalized Fermat curves $(X,H)$ and their automorphism groups, showing that every triple $(S,N, G)$ as before is determined by a class of subgroups of $H$ that satisfy certain invariance property. This approach establishes a correspondence between topological equivalence classes and an appropriate quotient set. As an application, we specialize our results to the case $k$ prime and $m=2$, including algebraic models and isogeny decompositions of their Jacobian varieties. We then discuss some examples for the cases $n=3$ and $n=5$, which are interesting in their own right.

Stochastic variance reduced gradient (SVRG) is an accelerated version of stochastic gradient descent based on variance reduction, and is promising for solving large-scale inverse problems. In this work, we analyze SVRG and a regularized version that incorporates a priori knowledge of the problem, for solving linear inverse problems in Hilbert spaces. We prove that, with suitable constant step size schedules and regularity conditions, the regularized SVRG can achieve optimal convergence rates in terms of the noise level without any early stopping rules, and standard SVRG is also optimal for problems with nonsmooth solutions under a priori stopping rules. The analysis is based on an explicit error recursion and suitable prior estimates on the inner loop updates with respect to the anchor point. Numerical experiments are provided to complement the theoretical analysis.

We show that an oriented surface in $\mathbb{R}^4$ containing double point singularities induces a map between the Khovanov homology groups of its boundary links in a functorial way. As part of this work, the movie moves of Carter and Saito are extended to surfaces with double points.

A set $S$ of vertices in a graph $G$ is a dominating set of $G$ if every vertex not in $S$ is adjacent to a vertex in~$S$. An independent dominating set in $G$ is a dominating set of $G$ with the additional property that it is an independent set. The domination number, $\gamma(G)$, and the independent domination number, $i(G)$, are the minimum cardinalities among all dominating sets and independent dominating sets in $G$, respectively. By definition, $\gamma(G) \le i(G)$ for all graphs $G$. Let $G$ be a connected cubic graph of order~$n$. In 1996 Reed [Combin.\ Probab.\ Comput.\ 5 (1996), 277--295] proved a breakthrough result that $\gamma(G) \le \frac{3}{8}n$. We prove the stronger result that if $G$ is different from $K_{3,3}$ and the $5$-prism $C_5 \, \Box \, K_2$, then $i(G) \le \frac{3}{8}n$. This proves a known conjecture. The bound is tight in the sense that there are infinite families of connected cubic graphs that achieve equality in this bound.

In this paper we consider the problem of solving quantum field theories with time dependent interaction strengths. We show that the recently formulated framework [P. R. Pasnoori, Phys. Rev. B 112, L060409 (2025)], which is a generalization of the regular Bethe ansatz technique, provides the exact many-body wavefunction. In this framework, the time-dependent Schrodinger equation is reduced to a set of analytic difference equations and matrix difference equations, called the quantum Knizhnik-Zamolodchikov (qKZ) equations. The consistency of the solution gives rise to constraints on the time-dependent interaction strengths. For interaction strengths satisfying these constraints, the system is integrable, and the solution to the qKZ and the analytic difference equations provides the explicit form of the many-body wavefunction that satisfies the time-dependent Schrodinger equation. We provide a concrete example by considering the $SU(2)$ Gross-Neveu model with time dependent interaction strength. Using this framework we solve the model with the most general time-dependent interaction strength and obtain the explicit form of the wave function.

We introduce the cut finite element method in the language of finite element exterior calculus, by formulating a stabilisation -- for any form degree -- that makes the method robust with respect to the position of the interface relative to the mesh. We prove that the $L^2$-norm on the physical domain augmented with this stabilisation is uniformly equivalent to the $L^2$-norm on the ``active'' mesh that contains all the degrees of freedom of the finite element space (including those external to the physical domain). We show how this CutFEEC method can be applied to discretize the Hodge Laplace equations on an unfitted mesh, in any dimension and any topology. A numerical illustration is provided involving a conforming finite element space of $H^{\text{curl}}$ posed on a filled torus, with convergence and condition number scaling independent of the position of the boundary with respect to the background mesh.

In a recent work, we presented the reduced Jacobian method (RJM) as an extension of Wolfe's reduced gradient method to multicriteria (multiobjective) optimization problems dealing with linear constraints. This approach reveals that using a reduction technique of the Jacobian matrix of the objective avoids scalarization. In the present work, we intend to generalize RJM to handle nonlinear constraints too. In fact, we propose a generalized reduced Jacobian (GRJ) method that extends Abadie-Carpentier's approach for single-objective programs. To this end, we adopt a global reduction strategy based on the fundamental theorem of implicit functions. In this perspective, only a reduced descent direction common to all the criteria is computed by solving a simple convex program. After establishing an Armijo-type line search condition that ensures feasibility, the resulting algorithm is shown to be globally convergent, under mild assumptions, to a Pareto critical (KKT-stationary) point. Finally, experimental results are presented, including comparisons with other deterministic and evolutionary approaches.

We continue the study of the level-set percolation of the discrete Gaussian free field (GFF) on regular trees in the critical regime, initiated in arXiv:2302.02753. First, we derive a sharp asymptotic estimate for the probability that the connected component of the critical level set containing the root of the tree reaches generation $n$. In particular, we show that the one-arm exponent satisfies $\rho =1$. Next, we establish a Yaglom-type limit theorem for the values of the GFF at generation $n$ within this component. Finally, we show that, after a correct rescaling, this component conditioned on reaching generation $n$ converges, as $n\to\infty$, to Aldous' continuum random tree.

We prove that a log surface has only finitely many weakly log canonical projective models with klt singularities up to log isomorphism, by reducing the problem to the boundedness of their polarization.

We introduce the $\Sigma^*$-invariant of a group of finite type, which is defined to be the subset of non-zero characters $\chi \in \mathrm H^1(G;\mathbb R)$ with vanishing associated top-dimensional Novikov cohomology. We prove an analogue of Sikorav's Theorem for this invariant, namely that $\mathrm{cd}(\ker \chi) = \mathrm{cd}(G) - 1$ if and only if $\pm \chi \in \Sigma^*(G)$ for integral characters $\chi$. This implies that cohomological dimension drop is an open property among integral characters. We also study the cohomological dimension of arbitrary co-Abelian subgroups. The techniques yield a short new proof of Ranicki's criterion for finite domination of infinite cyclic covers, and in a different direction, we prove that the algebra of affiliated operators $\mathcal U(G)$ of a RFRS group $G$ has weak dimension at most one if and only if $G$ is an iterated (cyclic or finite) extension of a free group.

We study the accuracy of a class of methods to compute the Inverse Laplace Transform, the so-called \emph{Abate--Whitt methods} [Abate, Whitt 2006], which are based on a linear combination of evaluations of $\widehat{f}$ in a few points. We provide error bounds which relate the accuracy of a method to the rational approximation of the exponential function. We specialize our analysis to applications in queuing theory, a field in which Abate--Whitt methods are often used; in particular, we study phase-type distributions and Markov-modulated fluid models (or \emph{fluid queues}). We use a recently developed algorithm for rational approximation, the AAA algorithm [Nakatsukasa, Sète, Trefethen 2018], to produce a new family of methods, which we call TAME. The parameters of these methods are constructed depending on a function-specific domain $\Omega$; we provide a quasi-optimal choice for certain families of functions. We discuss numerical issues related to floating-point computation, and we validate our results through numerical experiments which show that the new methods require significantly fewer function evaluations to achieve an accuracy that is comparable (or better) to that of the classical methods.

We prove that for any $k \ge 3$, every $k$-uniform hypergraph on $n$ vertices contains at most $n - o(n)$ different sizes of cliques (maximal complete subgraphs). In particular, the 3-uniform case answers a question of Erdős.

We propose a semismooth Newton-based augmented Lagrangian method for reconstructing sparse sources in inverse acoustic scattering problems. The semismooth Newton method can be iterated in the space of measurements instead of the unknown source to be reconstructed. It is highly efficient, especially when the measurement data is much less than the acoustic source. The source can be calculated from Fenchel-Rockafellar duality theory. We can obtain lots of acceleration and leverage the computational cost. The numerical examples show the high efficiency of the proposed semismooth Newton-based methods.

We study finite-time blow-up for the one-dimensional nonlinear wave equation with a quadratic time-derivative nonlinearity, \[ u_{tt}-u_{xx}=(u_t)^2,\qquad (x,t)\in\mathbb R\times[0,T). \] Building on the work of Ghoul, Liu, and Masmoudi \cite{ghoul2025blow} on the spatial-derivative analogue, we establish the non-existence of smooth, exact self-similar blow-up profiles. Instead we construct an explicit family of \emph{generalised self-similar} solutions, bifurcating from the ODE blow-up, that are smooth within the past light cone and exhibit type-I blow-up at a prescribed point \((x_0,T)\). We further prove asymptotic stability of these profiles under small perturbations in the energy topology. In particular, these profiles verify that the spatially homogeneous ODE blow-up is not asymptotically stable.

Polynomial preconditioning is an important tool in solving large linear systems and eigenvalue problems. A polynomial from GMRES can be used to precondition restarted GMRES and restarted Arnoldi. Here we give methods for indefinite matrices that make polynomial preconditioning more generally applicable. The new techniques include balancing the polynomial so that it produces a definite spectrum. Then a stability approach is given that is specialized for the indefinite case. Also, very complex spectra are examined. Then convergence estimates are given for polynomial preconditioning of real, indefinite spectra. Finally, tests are preformed of finding interior eigenvalues.

We provide, for any regular uncountable cardinal $\kappa$, a new argument for Pincus' result on the consistency of $\ZF$ with the higher dependent choice principle $\DC_{<\kappa}$ and the ordering principle in the presence of a failure of the axiom of choice. We also generalise his methods and obtain these consistency results in a larger class of models.

Model selection criteria are one of the most important tools in statistics. Proofs showing a model selection criterion is asymptotically optimal are tailored to the type of model (linear regression, quantile regression, penalized regression, etc.), the estimation method (linear smoothers, maximum likelihood, generalized method of moments, etc.), the type of data (i.i.d., dependent, high dimensional, etc.), and the type of model selection criterion. Moreover, assumptions are often restrictive and unrealistic making it a slow and winding process for researchers to determine if a model selection criterion is selecting an optimal model. This paper provides general proofs showing asymptotic optimality for a wide range of model selection criteria under general conditions. This paper not only asymptotically justifies model selection criteria for most situations, but it also unifies and extends a range of previously disparate results.

In this article, we study the existence and distribution of elements in finite field extensions with prescribed traces in several intermediate extensions that are also either normal or primitive normal. In the former case, we fully characterize the conditions under which such elements exist and provide an explicit enumeration of these elements. In the latter case we provide asymptotic results.

We analyze by density evolution the asymptotic performance of rate-adaptive MacKay-Neal (MN) code ensembles, where the inner code is a protograph spatially coupled (SC) low-density parity-check code. By resorting to a suitably-defined parallel channel model, we compute belief propagation decoding thresholds, showing that SC MN code ensembles can perform within 0.15 dB from the binary-input additive white Gaussian noise capacity over the full [0,1] rate range.

Rate-adaptive MacKay-Neal (MN) codes based on protographs are analyzed. The code construction employs an outer distribution matcher (DM) to adapt the rate of the scheme. The DM is coupled with an inner protograph-based low-density parity-check (LDPC) code. The performance achievable by the resulting code structure, that is nonlinear, is studied by means of an equivalent communication model that reduces the problem to the analysis of the inner (linear) LDPC code with transmission that takes place in parallel over the communication channel, and over a suitably defined binary symmetric channel. A density evolution analysis of protograph MN code ensembles is outlined, and it is complemented by an error floor analysis that relies on the derivation of the average input-output weight distribution of the inner LDPC code ensemble. Conditions on the shape of the normalized logarithmic asymptotic input-output weight distribution are defined, which allow discarding code ensembles with bad error floor properties during the code design phase. Examples of code designs are provided, showing how the use of a single LDPC code ensemble allows operating within 1 dB from the Shannon limit over a wide range of code rates, where the code rate is selected by tuning the DM parameters. By enabling rate flexibility with a constant blocklength, and with a fixed LDPC code as inner code, the construction provides an appealing solution for very high-throughput wireless (optical) links that employ binary-input modulations.

Given any vertex operator algebra $ V $ with an automorphism $ g $, we derive a Jacobi identity for an intertwining operator $ \mathcal{Y} $ of type $ \left( \begin{smallmatrix} W_3\\ W_1 \, W_2 \end{smallmatrix}\right) $ when $ W_1 $ is an untwisted $ V $-module, and $ W_2 $ and $ W_3 $ are $ g $-twisted $ V $-modules. We say such an intertwining operator is of $\left(\!\begin{smallmatrix} g\\ 1 \ g \end{smallmatrix}\!\right)$-type. Using the Jacobi identity, we obtain homogeneous linear differential equations satisfied by the multi-series $ \langle w_0, \mathcal{Y}_1(w_1,z_1) \cdots \mathcal{Y}_N(w_N,z_N) w_{N+1} \rangle $ when $ \mathcal{Y}_j $ are of $\left(\!\begin{smallmatrix} g\\ 1 \ g \end{smallmatrix}\!\right)$-type and the modules are $ C_1 $-cofinite and discretely graded. In the special case that $ V $ is an affine vertex operator algebra, we derive the ``twisted KZ equations" and show that its solutions have regular singularities at certain prescribed points when $ g $ has finite order. When $ V $ is general and $ g $ has finite order, we use the theory of regular singular points to prove that the multi-series $ \langle w_0, \mathcal{Y}_1(w_1,z_1) \cdots \mathcal{Y}_N(w_N,z_N) w_{N+1} \rangle $ converges absolutely to a multivalued analytic function when $ |z_1| > \cdots > |z_N| > 0 $ and analytically extends to the region $ z_i, z_i - z_j \neq 0 $. Furthermore, when $ N = 2 $, we show that these multivalued functions have regular singularities at certain prescribed points.

We show that any closed immersed curve in $\mathbb R^n$ with a one-to-one convex projection onto some $2$-plane develops a Type~I singularity and becomes asymptotically circular under Curve Shortening flow in $\mathbb R^n$. As an application, we prove an analog of Huisken's conjecture for Curve Shortening flow in $\mathbb R^n$, showing that any closed immersed curve in $\mathbb R^n$ can be perturbed to a closed immersed curve in $\mathbb R^{n+2}$ which shrinks to a round point under Curve Shortening flow.

Partial Information Decomposition (PID) was proposed by Williams and Beer in 2010 as a tool for analyzing fine-grained interactions between multiple random variables, and has since found numerous applications ranging from neuroscience to privacy. However, a unified theoretical framework remains elusive due to key conceptual and technical challenges. We identify and illustrate a crucial problem: PID violates the set-theoretic principle that the whole equals the sum of its parts (WESP). Through a counterexample in a three-variable system, we demonstrate how such violations naturally arise, revealing a fundamental limitation of current lattice-based PID frameworks. To address this issue, we introduce a new axiomatic framework, termed System Information Decomposition (SID), specifically tailored for three-variable systems. SID resolves the WESP violation by redefining the summation rules of decomposed information atoms based on synergistic relationships. However, we further show that for systems with four or more variables, no partial summation approach within the existing lattice-based structures can fully eliminate WESP inconsistencies. Our results thus highlight the inherent inadequacy of (antichain) lattice-based decompositions for general multivariate systems.

The classical Zarankiewicz problem, which concerns the maximum number of edges in a bipartite graph without a forbidden complete bipartite subgraph, motivates a direct analogue for hypergraphs. Let $K_{s_1,\ldots, s_r}$ be the complete $r$-partite $r$-graph such that the $i$-th part has $s_i$ vertices. We say an $r$-partite $r$-graph $H=H(V_1,\ldots,V_r)$ contains an ordered $K_{s_1,\ldots, s_r}$ if $K_{s_1,\ldots, s_r}$ is a subgraph of $H$ and the set of size $s_i$ vertices is embedded in $V_i$. The Zarankiewicz number for $r$-graph, denoted by $z(m_1, \ldots, m_{r}; s_1,, \ldots,s_{r})$, is the maximum number of edges of the $r$-partite $r$-graph whose $i$-th part has $m_i$ vertices and does not contain an ordered $K_{s_1,\ldots, s_r}$. In this paper, we show that $$z(m_1,m_2, \cdots, m_{r-1},n ; s_1,s_2, \cdots,s_{r-1}, t)=\Theta\left(m_1m_2\cdots m_{r-1} n^{1-1 / s_1s_2\cdots s_{r-1}}\right)$$ for a range of parameters. This extends a result of Conlon [Math. Proc. Camb. Philos. Soc. (2022)].

Two approaches are presented for computing upper bounds on Lyapunov exponents and their sums, and on Lyapunov dimension, among all trajectories of a dynamical system governed by ordinary differential equations. The first approach expresses a sum of Lyapunov exponents as a time average in an augmented dynamical system and then applies methods for bounding time averages. This generalizes the work of Oeri & Goluskin (Nonlinearity 36:5378-5400, 2023), who bounded the single leading Lyapunov exponent. The second approach considers a different augmented dynamical system, where bounds on sums of Lyapunov exponents are implied by stability of certain sets, and such stability is verified using Lyapunov function methods. Both of our approaches also can be adapted to directly compute bounds on Lyapunov dimension, which in turn implies bounds on the fractal dimension of a global attractor. For systems of ordinary differential equations with polynomial right-hand sides, all of our bounding formulations lead to polynomial optimization problems with sum-of-squares constraints. These sum-of-squares problems can be solved computationally for any particular system to yield numerical bounds, provided the number of variables and degree of polynomials is not prohibitive. Most of our upper bounds are proven to be sharp under relatively weak assumptions. In the case of the polynomial optimization problems, sharpness means that upper bounds converge to the exact values as polynomial degrees are raised. Computational examples demonstrate upper bounds that are sharp to several digits, including for a six-dimensional dynamical system where sums of Lyapunov exponents are maximized on periodic orbits.

We prove the stability and global convergence of a coupled actor-critic gradient flow for infinite-horizon and entropy-regularised Markov decision processes (MDPs) in continuous state and action space with linear function approximation under Q-function realisability. We consider a version of the actor critic gradient flow where the critic is updated using temporal difference (TD) learning while the policy is updated using a policy mirror descent method on a separate timescale. We demonstrate stability and exponential convergence of the actor critic flow to the optimal policy. Finally, we address the interplay of the timescale separation and entropy regularisation and its effect on stability and convergence.

We complete our study of the three dimensional Ginzburg--Landau functional with magnetic field, in the asymptotic regime of a small inverse Ginzburg--Landau parameter $\varepsilon$, and near the first critical field $H_{c_1}$ for which the first vortex filaments appear in energy minimizers. Under a nondegeneracy condition, we show a next order asymptotic expansion of $H_{c_1}$ as $\varepsilon \to 0$, and exhibit a sequence of transitions, with vortex lines appearing one by one as the intensity of the applied magnetic field is increased: passing $H_{c_1}$ there is one vortex, then increasing $H_{c_1}$ by an increment of order $\log |\log\varepsilon|$ a second vortex line appears, etc. These vortex lines accumulate near a special curve $\Gamma_0$, solution to an isoflux problem. We derive a next order energy that the vortex lines must minimize in the asymptotic limit, after a suitable horizontal blow-up around $\Gamma_0$. This energy is the sum of terms where penalizations of the length of the lines, logarithmic repulsion between the lines and magnetic confinement near $\Gamma_0$ compete. This elucidates the shape of vortex lines in superconductors.

In many applications, one seeks to approximate integration against a positive measure of interest by a positive discrete measure: a numerical quadrature rule with positive weights. One common desired discretization property is moment preservation over a finite dimensional function space, e.g., bounded-degree polynomials. Carathéodory's theorem asserts that if there is any finitely supported quadrature rule with more nodes than the dimension of the given function space, one can form a smaller (and hence more efficient) positive, nested, quadrature rule that preserves the moments of the original rule. We describe an efficient streaming procedure for Carathéodory-Steinitz pruning, a numerical procedure that implements Carathéodory's theorem for this measure compression. The new algorithm makes use of Givens rotations and on-demand storage of arrays to successfully prune very large rules whose storage complexity only depends on the dimension of the function space. This approach improves on a naive implementation of Carathéodory-Steinitz pruning whose runtime and storage complexity are quadratic and linear, respectively, in the size of the original measure. We additionally prove mathematical stability properties of our method with respect to a set of admissible, total-variation perturbations of the original measure. Our method is compared to two alternate approaches with larger storage requirements: non-negative least squares and linear programming, and we demonstrate comparable runtimes, with improved stability and storage robustness. Finally, we demonstrate practical usage of this algorithm to generate quadrature for discontinous Galerkin finite element simulations on cut-cell meshes.

Kernel functions are frequently encountered in differential equations and machine learning applications. In this work, we study the rank of matrices arising out of the kernel function $K: X \times Y \mapsto \mathbb{R}$, where the sets $X, Y \in \mathbb{R}^d$ are hypercubes that share a boundary. The main contribution of this work is the analysis of the rank of such matrices where the particles (sources/targets) are arbitrarily distributed within these hypercubes. To our knowledge, this is the first work to formally investigate the rank of such matrices for an arbitrary distribution of particles. We model the arbitrary distribution of particles to arise from an underlying random distribution and obtain bounds on the expected rank and variance of the rank of the kernel matrix corresponding to various neighbor interactions. These bounds are useful for understanding the performance and complexity of hierarchical matrix algorithms (especially hierarchical matrices satisfying the weak-admissibility criterion) for an arbitrary distribution of particles. We also present numerical experiments in one-, two-, and three-dimensions, showing the expected rank growth and variance of the rank for different types of interactions. The numerical results, not surprisingly, align with our theoretical predictions.

We present a broad family of high-order finite element algorithms for simulating the flow of electroneutral electrolytes. The governing partial differential equations that we solve are the electroneutral Navier-Stokes-Onsager-Stefan-Maxwell (NSOSM) equations, which model momentum transport, multicomponent diffusion and electrical effects within the electrolyte. Our algorithms can be applied in the steady and transient settings, in two and three spatial dimensions, and under a variety of boundary conditions. Moreover, we allow for the material parameters (e.g. viscosity, diffusivities, thermodynamic factors and density) to be solution-dependent and thermodynamically non-ideal. The flexibility of our approach requires us to address subtleties that arise in the governing equations due to the interplay between boundary conditions and the equation of state. We demonstrate the algorithms in various physical configurations, including (i) electrolyte flow around a microfluidic rotating disk electrode and (ii) the flow in a Hull cell of a cosolvent electrolyte mixture used in lithium-ion batteries.

In many applications of mathematical optimization, one may wish to optimize an objective function without access to its derivatives. These situations call for derivative-free optimization (DFO) methods. Among the most successful approaches in practice are model-based trust-region methods, such as those pioneered by M.J.D Powell. While relatively complex to implement, these methods are now available in standard scientific computing platforms, including MATLAB and SciPy. However, theoretical analysis of their computational complexity lags behind practice. In particular, it is important to bound the number of function evaluations required to achieve a desired level of accuracy. In this paper we systematically derive complexity bounds for classical model-based trust-region methods and their modern variations. We establish, for the first time, that these methods can have the same worst case complexity than any other known DFO method.

We define a weighted analog for the multidimensional Catalan numbers, obtain matrix-based recurrences for some of them, and give conditions under which they are periodic. Building on this framework, we introduce two new sequences of triangular arrays: the first one enumerates the $k$-dimensional Balanced ballot paths of exact height $s$; the second one is a new multidimensional generalization of the Narayana numbers, which count the number of Balanced ballot paths with exactly $p$ peaks.

In this work we present two new families of multirate time step adaptivity controllers, that are designed to work with embedded multirate infinitesimal (MRI) time integration methods for adapting time steps when solving problems with multiple time scales. We compare these controllers against competing approaches on two benchmark problems and see that they offer dramatically improved performance and flexibility, with each proposed family excelling on different types of multirate applications. The combination of embedded MRI methods and the proposed controllers enable adaptive simulations of problems with a potentially arbitrary number of time scales, achieving high accuracy while maintaining low computational cost. Additionally, we introduce a new set of embeddings for the family of explicit multirate exponential Runge--Kutta (MERK) methods of orders 2 through 5, resulting in the first-ever fifth-order embedded MRI method. Finally, we compare the performance of a wide range of embedded MRI methods on our benchmark problems to provide guidance on how to select an appropriate MRI method and multirate controller.

In this paper, we consider the properties of finite groups that are witnessed by group invariants arising in the context of Dijkgraaf--Witten theory, a topological quantum field theory, as invariants of surfaces. These invariants can be considered generalizations of the commuting probability, an invariant that has been well studied in the group theory literature.

Control problems frequently arise in scientific and industrial applications, where the objective is to steer a dynamical system from an initial state to a desired target state. Recent advances in deep learning and automatic differentiation have made applying these methods to control problems increasingly practical. In this paper, we examine the use of neural networks and modern machine-learning libraries to parameterize control inputs across discrete-time and continuous-time systems, as well as deterministic and stochastic dynamics. We highlight applications in multiple domains, including biology, engineering, physics, and medicine. For continuous-time dynamical systems, neural ordinary differential equations (neural ODEs) offer a useful approach to parameterizing control inputs. For discrete-time systems, we show how custom control-input parameterizations can be implemented and optimized using automatic-differentiation methods. Overall, the methods presented provide practical solutions for control tasks that are computationally demanding or analytically intractable, making them valuable for complex real-world applications.

We developed a low-energy model that can be used at any time to describe the dynamics of DNA bubbles at temperatures below the melting point. The Schrödinger equation associated with this problem is solved in imaginary time with a quantum Coulomb potential, and we obtain an approximate expression for its more general physical solution as a linear combination of the states whose energies are close to the lower bound energy. We can then determine the probability density, the first-passage time density, and the correlation functions in terms of Bessel functions. Our findings are consistent with results obtained directly from the Fokker-Planck equation. Comparisons with the Gamma and Diffusion models are discussed.

Many of the most fundamental observables | position, momentum, phase-point, and spin-direction | cannot be measured by an instrument that obeys the orthogonal projection postulate. Continuous-in-time measurements provide the missing theoretical framework to make sense of such observables. The elements of the time-dependent instrument define a group called the \emph{instrumental group} (IG). Relative to the IG, all of the time-dependence is contained in a certain function called the \emph{Kraus-operator density} (KOD), which evolves according to a classical Kolmogorov equation. Unlike the Lindblad master equation, the KOD Kolmogorov equation is a direct expression of how the elements of the instrument (not just the total channel) evolve. Shifting from continuous measurement to sequential measurements more generally, the structure of combining instruments in sequence is shown to correspond to the convolution of their KODs. This convolution promotes the IG to an \emph{involutive Banach algebra} (a structure that goes all the way back to the origins of POVM and C*-algebra theory) which will be called the \emph{instrumental group algebra} (IGA). The IGA is the true home of the KOD, similar to how the dual of a von Neumann algebra is the home of the density operator. Operators on the IGA, which play the same role for KODs as superoperators play for density operators, are called \emph{ultraoperators} and various examples are discussed. Certain ultraoperator-superoperator intertwining relations are considered, including the relation between the KOD Kolmogorov equation and the Lindblad master equation. The IGA is also shown to have actually two involutions: one respected by the convolution ultraoperators and the other by the quantum channel superoperators. Finally, the KOD Kolmogorov generators are derived for jump processes and more general diffusive processes.

Existing or planned power grids need to evaluate survivability under extreme events, like a number of peak load overloading conditions, which could possibly cause system collapses (i.e. blackouts). For realistic extreme events that are correlated or share similar patterns, it is reasonable to expect that the dominant vulnerability or failure sources behind them share the same locations but with different severity. Early warning diagnosis that proactively identifies the key vulnerabilities responsible for a number of system collapses of interest can significantly enhance resilience. This paper proposes a multi-period sparse optimization method, enabling the discovery of {persistent failure sources} across a sequence of collapsed systems with increasing system stress, such as rising demand or worsening contingencies. This work defines persistency and efficiently integrates persistency constraints to capture the ``hidden'' evolving vulnerabilities. Circuit-theory based power flow formulations and circuit-inspired optimization heuristics are used to facilitate the scalability of the method. Experiments on benchmark systems show that the method reliably tracks persistent vulnerability locations under increasing load stress, and solves with scalability to large systems ({on average} taking {around} 200 s per scenario on 2000+ bus systems).

As the scope of Computational Fluid Dynamics (CFD) grows to encompass ever larger problem scales, so does the interest in whether quantum computing can provide an advantage. In recent years, Quantum Lattice Gas Automata (QLGA) and Quantum Lattice Boltzmann Methods (QLBM) have emerged as promising candidates for quantum-native implementations of CFD solvers. Though the progress in developing QLGA and QLBM algorithms has been significant, it has largely focused on the development of models rather than applications. As a result, the zoo of QLGA and QLBM algorithms has grown to target several equations and to support many extensions, but the practical use of these models is largely limited to quantum state tomography and observable measurement. This limitation is crucial in practice, because unless very specific criteria are met, such measurements may cancel out any potential quantum advantage. In this paper, we propose an application based on discrete optimization and quantum search, which circumvents flow field measurement altogether. We propose methods for simulating many different lattice configurations simultaneously and describe how the usage of amplitude estimation and quantum search can provide an asymptotic quantum advantage. Throughout the paper, we provide detailed complexity analyses of gate-level implementations of our circuits and consider the benefits and costs of several encodings.

We study the expressivity of sparse maxout networks, where each neuron takes a fixed number of inputs from the previous layer and employs a, possibly multi-argument, maxout activation. This setting captures key characteristics of convolutional or graph neural networks. We establish a duality between functions computable by such networks and a class of virtual polytopes, linking their geometry to questions of network expressivity. In particular, we derive a tight bound on the dimension of the associated polytopes, which serves as the central tool for our analysis. Building on this, we construct a sequence of depth hierarchies. While sufficiently deep sparse maxout networks are universal, we prove that if the required depth is not reached, width alone cannot compensate for the sparsity of a fixed indegree constraint.

We develop a framework for analyzing the training and learning rate dynamics on a variety of high- dimensional optimization problems trained using one-pass stochastic gradient descent (SGD) with data generated from multiple anisotropic classes. We give exact expressions for a large class of functions of the limiting dynamics, including the risk and the overlap with the true signal, in terms of a deterministic solution to a system of ODEs. We extend the existing theory of high-dimensional SGD dynamics to Gaussian-mixture data and a large (growing with the parameter size) number of classes. We then investigate in detail the effect of the anisotropic structure of the covariance of the data in the problems of binary logistic regression and least square loss. We study three cases: isotropic covariances, data covariance matrices with a large fraction of zero eigenvalues (denoted as the zero-one model), and covariance matrices with spectra following a power-law distribution. We show that there exists a structural phase transition. In particular, we demonstrate that, for the zero-one model and the power-law model with sufficiently large power, SGD tends to align more closely with values of the class mean that are projected onto the "clean directions" (i.e., directions of smaller variance). This is supported by both numerical simulations and analytical studies, which show the exact asymptotic behavior of the loss in the high-dimensional limit.

We study a two-sided market, wherein, price-sensitive heterogeneous customers and servers arrive and join their respective queues. A compatible customer-server pair can then be matched by the platform, at which point, they leave the system. Our objective is to design pricing and matching algorithms that maximize the platform's profit, while maintaining reasonable queue lengths. As the demand and supply curves governing the price-dependent arrival rates may not be known in practice, we design a novel online-learning-based pricing policy and establish its near-optimality. In particular, we prove a tradeoff among three performance metrics: $\tilde{O}(T^{1-\gamma})$ regret, $\tilde{O}(T^{\gamma/2})$ average queue length, and $\tilde{O}(T^{\gamma})$ maximum queue length for $\gamma \in (0, 1/6]$, significantly improving over existing results [1]. Moreover, barring the permissible range of $\gamma$, we show that this trade-off between regret and average queue length is optimal up to logarithmic factors under a class of policies, matching the optimal one as in [2] which assumes the demand and supply curves to be known. Our proposed policy has two noteworthy features: a dynamic component that optimizes the tradeoff between low regret and small queue lengths; and a probabilistic component that resolves the tension between obtaining useful samples for fast learning and maintaining small queue lengths.

The key purpose of this paper is to present Fourier method to model the stochastic time-change in this context of time-subordinated Brownian motion models. We review Gaussian Variance-Mean mixtures and time-subordinated models with a key example of the Gamma process. A non-parametric characteristic function decomposition of subordinated Brownian motion is presented. This allows one to characterise and study the stochastic time-change directly from the full process. Finally we provide an example empirical decomposition of S$\&$P log-returns. We explore the Variance Gamma process as a key example throughout.

Shadow molecular dynamics provide an efficient and stable atomistic simulation framework for flexible charge models with long-range electrostatic interactions. While previous implementations have been limited to atomic monopole charge distributions, we extend this approach to flexible multipole models. We derive detailed expressions for the shadow energy functions, potentials, and force terms, explicitly incorporating monopole-monopole, dipole-monopole, and dipole-dipole interactions. In our formulation, both atomic monopoles and atomic dipoles are treated as extended dynamical variables alongside the propagation of the nuclear degrees of freedom. We demonstrate that introducing the additional dipole degrees of freedom preserves the stability and accuracy previously seen in monopole-only shadow molecular dynamics simulations. Additionally, we present a shadow molecular dynamics scheme where the monopole charges are held fixed while the dipoles remain flexible. Our extended shadow dynamics provide a framework for stable, computationally efficient, and versatile molecular dynamics simulations involving long-range interactions between flexible multipoles. This is of particular interest in combination with modern artificial intelligence and machine learning techniques, which are increasingly used to develop physics-informed and data-driven foundation models for atomistic simulations. These models aim to provide transferable, high-accuracy representations of atomic interactions that are applicable across diverse sets of molecular systems, which requires accurate treatment of long-range charge interactions.

Pinching-antenna systems have emerged as a novel and transformative flexible-antenna architecture for next-generation wireless networks. They offer unprecedented flexibility and spatial reconfigurability by enabling dynamic positioning and activation of radiating elements along a signal-guiding medium (e.g., dielectric waveguides), which is not possible with conventional fixed antenna systems. In this paper, we introduce the concept of generalized pinching antenna systems, which retain the core principle of creating localized radiation points on demand, but can be physically realized in a variety of settings. These include implementations based on dielectric waveguides, leaky coaxial cables, surface-wave guiding structures, and other types of media, employing different feeding methods and activation mechanisms (e.g., mechanical, electronic, or hybrid). Despite differences in their physical realizations, they all share the same inherent ability to form, reposition, or deactivate radiation sites as needed, enabling user-centric and dynamic coverage. We first describe the underlying physical mechanisms of representative generalized pinching-antenna realizations and their associated wireless channel models, highlighting their unique propagation and reconfigurability characteristics compared with conventional antennas. Then, we review several representative pinching-antenna system architectures, ranging from single- to multiple-waveguide configurations, and discuss advanced design strategies tailored to these flexible deployments. Furthermore, we examine their integration with emerging wireless technologies to enable synergistic, user-centric solutions. Finally, we identify key open research challenges and outline future directions, charting a pathway toward the practical deployment of generalized pinching antennas in next-generation wireless networks.

It has been known for many years that, in Yang-Mills theories with $\mathcal{N}=4,2,2^*$ supersymmetry, certain nontrivial supersymmetric Wilson loops exist with v.e.v. either trivial or computable by localization that arises from a cohomological field theory, which also computes the nonperturbative prepotential in $\mathcal{N}=2,2^*$ theories. Moreover, some years ago it has been argued that, in analogy with the supersymmetric case, certain nontrivial twistor Wilson loops with trivial v.e.v. to the leading large-$N$ order exist in pure SU($N$) Yang-Mills theory and are computed, to the leading large-$N$ order, by a topological field/string theory that, to the next-to-leading $\frac{1}{N}$ order, conjecturally captures nonperturbative information on the glueball spectrum and glueball one-loop effective action as well. In fact, independently of the above, it has also been claimed that "every gauge theory with a mass gap should contain a possibly trivial topological field theory in the infrared", so that the aforementioned twistor Wilson loops realize a stronger version of this idea, as they have trivial v.e.v. at all energy scales and not only in the infrared. In the present paper, we provide a detailed proof of the triviality of the v.e.v. of twistor Wilson loops at the leading large-$N$ order in Yang-Mills theory that has previously been only sketched, opening the way to further developments.

We characterize the local instability of pressureless Friedmann spacetimes to radial perturbation at the Big Bang. The analysis is based on a formulation of the Einstein-Euler equations in self-similar variables $(t,\xi)$, with $\xi=r/t$, conceived to realize the critical ($k=0$) Friedmann spacetime as a stationary solution whose character as an unstable saddle rest point $SM$ is determined via an expansion of smooth solutions in even powers of $\xi$. The eigenvalues of $SM$ imply the $k\neq0$ Friedmann spacetimes are unstable solutions within the unstable manifold of $SM$. We prove that all solutions smooth at the center of symmetry agree with a Friedmann spacetime at leading order in $\xi$, and with an eye toward Cosmology, we focus on $\mathcal{F}$, the set of solutions which agree with a $k<0$ Friedmann spacetime at leading order, providing the maximal family into which generic underdense radial perturbations of the unstable critical Friedmann spacetime will evolve. We prove solutions in $\mathcal{F}$ generically accelerate away from Friedmann spacetimes at intermediate times but decay back to the same leading order Friedmann spacetime asymptotically as $t\to\infty$. Thus instabilities inherent in the Einstein-Euler equations provide a natural mechanism for an accelerated expansion without recourse to a cosmological constant or dark energy.

This paper presents an initial investigation into the combination of integrated sensing and communication (ISAC) and massive communication, both of which are largely regarded as key scenarios in sixth-generation (6G) wireless networks. Specifically, we consider a cell-free network comprising a large number of users, multiple targets, and distributed base stations (BSs). In each time slot, a random subset of users becomes active, transmitting pilot signals that can be scattered by the targets before reaching the BSs. Unlike conventional massive random access schemes, where the primary objectives are device activity detection and channel estimation, our framework also enables target localization by leveraging the multipath propagation effects introduced by the targets. However, due to the intricate dependency between user channels and target locations, characterizing the posterior distribution required for minimum mean-square error (MMSE) estimation presents significant computational challenges. To handle this problem, we propose a hybrid message passing-based framework that incorporates multiple approximations to mitigate computational complexity. Numerical results demonstrate that the proposed approach achieves high-accuracy device activity detection, channel estimation, and target localization simultaneously, validating the feasibility of embedding localization functionality into massive communication systems for future 6G networks.

We consider a load balancing system consisting of $n$ single-server queues working in parallel, with heterogeneous service rates. Jobs arrive to a central dispatcher, which has to dispatch them to one of the queues immediately upon arrival. For this setting, we consider a broad family of policies where the dispatcher can only access the queue lengths sporadically, every $T$ units of time. We assume that the dispatching decisions are made based only on the order of the scaled queue lengths at the last time that the queues were accessed, and on the processing rate of each server. For these general policies, we provide easily verifiable necessary and sufficient conditions for the stability of the system, and sufficient conditions for heavy-traffic delay optimality. We also show that, in heavy-traffic, the queue length converges in distribution to a scaled deterministic vector, where the scaling factor is an exponential random variable.

Volatility estimation is a central problem in financial econometrics, but becomes particularly challenging when jump activity is high, a phenomenon observed empirically in highly traded financial securities. In this paper, we revisit the problem of spot volatility estimation for an Itô semimartingale with jumps of unbounded variation. We construct truncated kernel-based estimators and debiased variants that extend the efficiency frontier for spot volatility estimation in terms of the jump activity index $Y$, raising the previous bound $Y<4/3$ to $Y<20/11$, thereby covering nearly the entire admissible range $Y<2$. Compared with earlier work, our approach attains smaller asymptotic variances through the use of unbounded kernels, is simpler to implement, and has broader applicability under more flexible model assumptions. A comprehensive simulation study confirms that our procedures substantially outperform competing methods in finite samples.

Accurate simulations of the flow in the human airway are essential for advancing diagnostic methods. Many existing computational studies rely on simplified geometries or turbulence models, limiting their simulation's ability to resolve flow features such shear-layer instabilities or secondary vortices. In this study, direct numerical simulations were performed for inspiratory flow through a detailed airway model which covers the nasal mask region to the 6th bronchial bifurcation. Simulations were conducted at two physiologically relevant \textsc{Reynolds} numbers with respect to the pharyngeal diameter, i.e., at Re_p=400 (resting) and Re_p=1200 (elevated breathing). These values characterize resting and moderately elevated breathing conditions. A lattice-Boltzmann method was employed to directly simulate the flow, i.e., no turbulence model was used. The flow field was examined across four anatomical regions: 1) the nasal cavity, 2) the naso- and oropharynx, 3) the laryngopharynx and larynx, and 4) the trachea and carinal bifurcation. The total pressure loss increased from 9.76 Pa at Re_p=400 to 41.93 Pa at Re_p=1200. The nasal cavity accounted for the majority of this loss for both Reynolds numbers, though its relative contribution decreased from 81.3% at Re_p=400 to 73.4% at Re_p=1200. At Re_p=1200, secondary vortices in the nasopharyngeal bend and turbulent shear-layers in the glottis jet enhanced the local pressure losses. In contrast, the carinal bifurcation mitigated upstream unsteadiness and stabilized the flow. A key outcome is the spatial correlation between the pressure loss and the onset of flow instabilities across the four regions. This yields a novel perspective on how the flow resistance and vortex dynamics vary with geometric changes and flow rate.

Recent work in the information sciences, especially informetrics and scientometrics, has made substantial contributions to the development of new metrics that eschew the intrinsic biases of citation metrics. This work has tended to employ either network scientific (topological) approaches to quantifying the disruptiveness of peer-reviewed research, or topic modeling approaches to quantifying conceptual novelty. We propose a combination of these approaches, investigating the prospect of topological data analysis (TDA), specifically persistent homology and mixup barcodes, as a means of understanding the negative space among document embeddings generated by topic models. Using top2vec, we embed documents and topics in n-dimensional space, we use persistent homology to identify holes in the embedding distribution, and then use mixup barcodes to determine which holes are being filled by a set of unobserved publications. In this case, the unobserved publications represent research that was published before or after the data used to train top2vec. We investigate the extent that negative embedding space represents missing context (older research) versus innovation space (newer research), and the extend that the documents that occupy this space represents integrations of the research topics on the periphery. Potential applications for this metric are discussed.

We present a geometric formulation of automatic differentiation (AD) using jet bundles and Weil algebras. Reverse-mode AD emerges as cotangent-pullback, while Taylor-mode corresponds to evaluation in a Weil algebra. From these principles, we derive concise statements on correctness, stability, and complexity: a functorial identity for reverse-mode, algebraic exactness of higher-order derivatives, and explicit bounds on truncation error. We further show that tensorized Weil algebras permit one-pass computation of all mixed derivatives with cost linear in the algebra dimension, avoiding the combinatorial blow-up of nested JVP/VJP schedules. This framework interprets AD theory through the lens of differential geometry and offers a foundation for developing structure-preserving differentiation methods in deep learning and scientific computing. Code and examples are available at this https URL.

Prange's information set algorithm is a decoding algorithm for arbitrary linear codes. It decodes corrupted codewords of any $\mathbb{F}_2$-linear code $C$ of message length $n$ up to relative error rate $O(\log n / n)$ in $\mathsf{poly}(n)$ time. We show that the error rate can be improved to $O((\log n)^2 / n)$, provided: (1) the decoder has access to a polynomial-length advice string that depends on $C$ only, and (2) $C$ is $n^{-\Omega(1)}$-balanced. As a consequence we improve the error tolerance in decoding random linear codes if inefficient preprocessing of the code is allowed. This reveals potential vulnerabilities in cryptographic applications of Learning Noisy Parities with low noise rate. Our main technical result is that the Hamming weight of $Hw$, where $H$ is a random sample of *short dual* codewords, measures the proximity of a word $w$ to the code in the regime of interest. Given such $H$ as advice, our algorithm corrects errors by locally minimizing this measure. We show that for most codes, the error rate tolerated by our decoder is asymptotically optimal among all algorithms whose decision is based on thresholding $Hw$ for an arbitrary polynomial-size advice matrix $H$.

This letter proposes a novel anti-interference communication method leveraging computational antennas, utilizing time averaging and 1-bit reconfigurable intelligent surfaces (RIS) to achieve robust signal modulation with minimal hardware complexity. We develop a communication model for computational antennas and propose an efficient signal processing algorithm optimized for temporal modulation. A USRP-based experimental platform is established to validate the approach under strong interference conditions (e.g., 5 dB jamming-to-signal ratio). Experimental results reveal up to an 80.9\% reduction in bit error rate (BER) and effective restoration of distorted images in transmission tests. Compared to conventional techniques like spread spectrum or frequency hopping, which require significant spectral resources, our method offers superior anti-interference performance without additional spectral overhead. This research provides valuable insights for radar detection, military communications, and next-generation wireless networks.

This paper develops a sensitivity analysis framework that transfers the average total treatment effect (ATTE) from source data with a fully observed network to target data whose network is completely unknown. The ATTE represents the average social impact of a policy that assigns the treatment to every individual in the dataset. We postulate a covariate-shift type assumption that both source and target datasets share the same conditional mean outcome. However, because the target network is unobserved, this assumption alone is not sufficient to pin down the ATTE for the target data. To address this issue, we consider a sensitivity analysis based on the uncertainty of the target network's degree distribution, where the extent of uncertainty is measured by the Wasserstein distance from a given reference degree distribution. We then construct bounds on the target ATTE using a linear programming-based estimator. The limiting distribution of the bound estimator is derived via the functional delta method, and we develop a wild bootstrap approach to approximate the distribution. As an empirical illustration, we revisit the social network experiment on farmers' weather insurance adoption in China by Cai et al. (2015).

We study the Tracy-Widom (TW) distribution $f_\beta(a)$ in the limit of large Dyson index $\beta \to +\infty$. This distribution describes the fluctuations of the rescaled largest eigenvalue $a_1$ of the Gaussian (alias Hermite) ensemble (G$\beta$E) of (infinitely) large random matrices. We show that, at large $\beta$, its probability density function takes the large deviation form $f_\beta(a) \sim e^{-\beta \Phi(a)}$. While the typical deviation of $a_1$ around its mean is Gaussian of variance $O(1/\beta)$, this large deviation form describes the probability of rare events with deviation $O(1)$, and governs the behavior of the higher cumulants. We obtain the rate function $\Phi(a)$ as a solution of a Painlevé II equation. We derive explicit formula for its large argument behavior, and for the lowest cumulants, up to order 4. We compute $\Phi(a)$ numerically for all $a$ and compare with exact numerical computations of the TW distribution at finite $\beta$. These results are obtained by applying saddle-point approximations to an associated problem of energy levels $E=-a$, for a random quantum Hamiltonian defined by the stochastic Airy operator (SAO). We employ two complementary approaches: (i) we use the optimal fluctuation method to find the most likely realization of the noise in the SAO, conditioned on its ground-state energy being $E$ (ii) we apply the weak-noise theory to the representation of the TW distribution in terms of a Ricatti diffusion process associated to the SAO. We extend our results to the full Airy point process $a_1>a_2>\dots$ which describes all edge eigenvalues of the G$\beta$E, and correspond to (minus) the higher energy levels of the SAO, obtaining large deviation forms for the marginal distribution of $a_i$, the joint distributions, and the gap distributions.

We investigate the asymmetric integrable turbulence and rogue waves (RWs) emerging from the modulation instability (MI) of plane waves for the DNLS equation. The \(n\)-th moments and ensemble-averaged kinetic and potential energy exhibit oscillatory convergence towards their steady-state values. Specifically, the amplitudes of oscillations for these indexes decay asymptotically with time as \(t^{-1.36}\), while the phase shifts demonstrate a nonlinear decay with a rate of \(t^{-0.78}\). The frequency of these oscillations is observed to be twice the maximum growth rate of MI. These oscillations can be classified into two distinct types: one is in phase with ensemble-averaged potential energy modulus $|\langle H_4\rangle|$, and the other is anti-phase. At the same time, this unity is also reflected in the wave-action spectrum \( S_k(t) \) for a given \( k \), the auto-correlation function \( g(x,t) \) for a given \( x \), as well as the PDF \( P(I,t) \). The critical feature of the turbulence is the wave-action spectrum, which follows a power-law distribution of \( |k+3|^{-\alpha} \) expect for $k=-3$. Unlike the NLS equation, the turbulence in the DNLS setting is asymmetric, primarily due to the asymmetry between the wave number of the plane wave from the MI and the perturbation wave number.. As the asymptotic peak value of \( S_k \) is observed at \( k = -3 \), the auto-correlation function exhibits a nonzero level as \( x \to \pm L/2 \). The PDF of the wave intensity asymptotically approaches the exponential distribution in an oscillatory manner. However, during the initial stage of the nonlinear phase, MI slightly increases the occurrence of RWs. This happens at the moments when the potential modulus is at its minimum, where the probability of RWs occurring in the range of \( I\in [12, 15] \) is significantly higher than in the asymptotic steady state.

The tt*-equation (topological-anti-topological fusion equation) was introduced by S. Cecotti and C. Vafa for describing massive deformation of supersymmetric conformal field theories. B. Dubrovin formulated the tt*-equation as a flat bundle, called tt*-structure. In this paper, we construct a tt*-structure for the quantum cohomology of the Grassmannian of complex \(k\)-plane and obtain global solutions to the tt*-equation, following the idea of Bourdeau. We give a precise mathematical formulation and a description of the solutions by using p.d.e. theory and the harmonic map theory developed by J. Dorfmeister, F. Pedit and H. Wu (the DPW method). Furthermore, we give an isomorphism between tt*-structure for the \(k\)-th exterior product of tt*-structure for the quantum cohomology of the complex projective space and the tt*-structure for the quantum cohomology of the Grassmannian.

Phase retrieval is a nonlinear inverse problem that arises in a wide range of imaging modalities, from electron microscopy to optical Fourier ptychography. Among various modalities, random phase retrieval stands out thanks to its strong theoretical guarantees and efficient reconstruction algorithms, although its applicability is hindered by prohibitive computational costs. In this paper, we propose the structured random models for phase retrieval, where we emulate a dense random matrix by a cascade of structured transforms and random diagonal matrices. We demonstrate that structured random models can achieve the same reconstruction performance as dense random models, with complexity reduced from quadratic to log-linear. Using a spectral method initialization followed by gradient descent, robust reconstruction is obtained at an oversampling ratio as low as 2.8. Moreover, we observe that the reconstruction performance is solely determined by the singular value distribution of the forward matrix. This class of models can directly be implemented with basic optical elements such as lenses and diffusers, paving the way for large-scale phase imaging with robust reconstruction guarantees.

This paper studies the error rate performance and low-complexity receiver design for zero-padded affine frequency division multiplexing (ZP-AFDM) systems. By exploiting the unique ZP-aided lower triangular structure of the time domain (TD) channel matrix, we propose {a novel low-complexity} minimum mean square error (MMSE) detector and {a} maximum ratio combining-based TD (MRC-TD) detector. Furthermore, the theoretical bit error rate (BER) {performance} of both MMSE and maximum likelihood detectors {is} analyzed. Simulation results demonstrate {that} the proposed detectors can achieve identical BER performance to that of {the conventional MMSE detector based on matrix inversion} while {enjoying significantly reduced complexity.}

Selecting the latent dimensions (ranks) in tensor factorization is a central challenge that often relies on heuristic methods. This paper introduces a rigorous approach to determine rank identifiability in probabilistic tensor models, based on prior predictive moment matching. We transform a set of moment matching conditions into a log-linear system of equations in terms of marginal moments, prior hyperparameters, and ranks; establishing an equivalence between rank identifiability and the solvability of such system. We apply this framework to four foundational tensor-models, demonstrating that the linear structure of the PARAFAC/CP model, the chain structure of the Tensor Train model, and the closed-loop structure of the Tensor Ring model yield solvable systems, making their ranks identifiable. In contrast, we prove that the symmetric topology of the Tucker model leads to an underdetermined system, rendering the ranks unidentifiable by this method. For the identifiable models, we derive explicit closed-form rank estimators based on the moments of observed data only. We empirically validate these estimators and evaluate the robustness of the proposal.

A bilateralist take on proof-theoretic semantics can be understood as demanding of a proof system to display not only rules giving the connectives' provability conditions but also their refutability conditions. On such a view, then, a system with two derivability relations is obtained, which can be quite naturally expressed in a proof system of natural deduction but which faces obstacles in a sequent calculus representation. Since in a sequent calculus there are two derivability relations inherent, one expressed by the sequent sign and one by the horizontal lines holding between sequents, in a truly bilateral calculus both need to be dualized. While dualizing the sequent sign is rather straightforwardly corresponding to dualizing the horizontal lines in natural deduction, dualizing the horizontal lines in sequent calculus, uncovers problems that, as will be argued in this paper, shed light on deeper conceptual issues concerning an imbalance between the notions of proof vs. refutation. The roots of this problem will be further analyzed and possible solutions on how to retain a bilaterally desired balance in our system are presented.

Given a family $\mathcal{F}$ of graphs, a graph is \emph{$\mathcal{F}$-subgraph-free} if it has no subgraph isomorphic to a member of $\mathcal{F}$. We present a fixed-parameter linear-time algorithm that decides whether a planar graph can be made $\mathcal{F}$-subgraph-free by deleting at most $k$ vertices or $k$ edges, where the parameters are $k$, $\lvert \mathcal{F} \rvert$, and the maximum number of vertices in a member of $\mathcal{F}$. The running time of our algorithm is double-exponential in the parameters, which is faster than the algorithm obtained by applying the first-order model checking result for graphs of bounded twin-width. To obtain this result, we develop a unified framework for designing algorithms for this problem on graphs with a ``product structure.'' Using this framework, we also design algorithms for other graph classes that generalize planar graphs. Specifically, the problem admits a fixed-parameter linear time algorithm on disk graphs of bounded local radius, and a fixed-parameter almost-linear time algorithm on graphs of bounded genus. Finally, we show that our result gives a tight fixed-parameter algorithm in the following sense: Even when $\mathcal{F}$ consists of a single graph $F$ and the input is restricted to planar graphs, it is unlikely to drop any parameters $k$ and $\lvert V(F) \rvert$ while preserving fixed-parameter tractability, unless the Exponential-Time Hypothesis fails.

We consider the quantitative description of a many-particle gas of interacting abelian anyons in the plane, confined in a trapping potential. If the anyons are modeled as bosons with a magnetic flux attachment, and if the total magnetic flux is small compared to the number of particles, then an average-field description becomes appropriate for the low-energy collective state of the gas. Namely, by means of a Hartree-Jastrow ansatz, we derive a two-parameter Chern-Simons-Schrödinger energy functional which extends the well-known Gross-Pitaevskii / nonlinear Schrödinger density functional theory to the magnetic (anyonic) self-interaction. One parameter determines the total number of self-generated magnetic flux units in the system, and the other the effective strength of spin-orbit self-interaction. This latter interaction can be either attractive/focusing or repulsive/defocusing, and depends both on the intrinsic spin-orbit interaction and the relative length scale of the flux profile of the anyons. Densities and energies of ground and excited states are studied analytically and numerically for a wide range of the parameters and align well with a sequence of exact nonlinear Landau levels describing Jackiw-Pi self-dual solitons. With increasing flux, counter-rotating vortices are formed, enhancing the stability of the gas against collapse. Apart from clarifying the relations between various different anyon models that have appeared in the literature, our analysis sheds considerable new light on the many-anyon spectral problem, and also exemplifies a novel supersymmetry-breaking phenomenon.

Increasing the batch size during training -- a ''batch ramp'' -- is a promising strategy to accelerate large language model pretraining. While for SGD, doubling the batch size can be equivalent to halving the learning rate, the optimal strategy for adaptive optimizers like Adam is less clear. As a result, any batch-ramp scheduling, if used at all, is typically tuned heuristically. This work develops a principled framework for batch-size scheduling and introduces Seesaw: whenever a standard scheduler would halve the learning rate, Seesaw instead multiplies it by $1/\sqrt{2}$ and doubles the batch size, preserving loss dynamics while reducing serial steps. Theoretically, we provide, to our knowledge, the first finite-sample proof of equivalence between learning-rate decay and batch-size ramp-up for SGD on noisy linear regression, and we extend this equivalence to normalized SGD, a tractable proxy for Adam, under a variance-dominated regime observed in practice. Empirically, on 150M/300M/600M-parameter models trained at Chinchilla scale using a constant (critical) batch size, Seesaw matches cosine decay at equal FLOPs while reducing wall-clock time by $\approx 36\%$, approaching the theoretical limit implied by our analysis.

Traditionally, the problem of apportioning the seats of a legislative body has been viewed as a one-shot process with no dynamic considerations. While this approach is reasonable for some settings, dynamic aspects play an important role in many others. We initiate the study of apportionment problems in an online setting. Specifically, we introduce a framework for proportional apportionment with no information about the future. In this model, time is discrete and there are $n$ parties that receive a certain share of the votes at each time step. An online algorithm needs to irrevocably assign a prescribed number of seats at each time, ensuring that each party receives its fractional share rounded up or down, and that the cumulative number of seats allocated to each party remains close to its cumulative share up to that time. We study deterministic and randomized online apportionment methods. For deterministic methods, we construct a family of adversarial instances that yield a lower bound, linear in $n$, on the worst-case deviation between the seats allocated to a party and its cumulative share. We show that this bound is best possible and is matched by a natural greedy method. As a consequence, a method guaranteeing that the cumulative number of seats assigned to each party up to any step equals its cumulative share rounded up or down (global quota) exists if and only if $n\leq 3$. Then, we turn to randomized allocations and show that, for $n\leq 3$, we can randomize over methods satisfying global quota with the additional guarantee that each party receives, in expectation, its proportional share in every step. Our proof is constructive: Any method satisfying these properties can be obtained from a flow on a recursively constructed network. We showcase the applicability of our results to obtain approximate solutions in the context of online dependent rounding procedures.

We propose an epidemic model for the spread of vector-borne diseases. The model, which is built extending the classical susceptible-infected-susceptible model, accounts for two populations -- humans and vectors -- and for cross-contagion between the two species, whereby humans become infected upon interaction with carrier vectors, and vectors become carriers after interaction with infected humans. We formulate the model as a system of ordinary differential equations and leverage monotone systems theory to rigorously characterize the epidemic dynamics. Specifically, we characterize the global asymptotic behavior of the disease, determining conditions for quick eradication of the disease (i.e., for which all trajectories converge to a disease-free equilibrium), or convergence to a (unique) endemic equilibrium. Then, we incorporate two control actions: namely, vector control and incentives to adopt protection measures. Using the derived mathematical tools, we assess the impact of these two control actions and determine the optimal control policy.

We consider an infinite balls-into-bins process with deletions where in each discrete step $t$ a coin is tossed as to whether, with probability $\beta(t) \in (0,1)$, a new ball is allocated using the Greedy[2] strategy (which places the ball in the lower loaded of two bins sampled uniformly at random) or, with remaining probability $1-\beta(t)$, a ball is deleted from a non-empty bin chosen uniformly at random. Let $n$ be the number of bins and $m(t)$ the total load at time $t$. We are interested in bounding the discrepancy $x_{\max}(t) - m(t)/n$ (current maximum load relative to current average) and the overload $x_{\max}(t) - m_{\max}(t)/n$ (current maximum load relative to highest average observed so far). We prove that at an arbitrarily chosen time $t$ the total number of balls above the average is $O(n)$ and that the discrepancy is $ O(\log(n))$. For the discrepancy, we provide a matching lower bound. Furthermore we prove that at an arbitrarily chosen time $t$ the overload is $\log\log(n)+O(1)$. For "good" insertion probability sequences (in which the average load of time intervals with polynomial length increases in expectation) we show that even the discrepancy is bounded by $\log\log(n)+O(1)$. One of our main analytical tools is a layered induction, as per [ABKU99]. Since our model allows for rather more general scenarios than what was previously considered, the formal analysis requires some extra ingredients as well, in particular a detailed potential analysis. Furthermore, we simplify the setup by applying probabilistic couplings to obtain certain "recovery" properties, which eliminate much of the need for intricate and careful conditioning elsewhere in the analysis.

Topological symmetries, invertible and otherwise, play a fundamental role in the investigation of quantum field theories. Despite their ubiquitous importance across a multitude of disciplines ranging from string theory to condensed matter physics, controlled realizations of models exhibiting these symmetries in physical systems are rare. Quantum simulators based on engineered solid-state devices provide a novel alternative to conventional condensed matter systems for realizing these models. In this work, eigenstates of impurity Hamiltonians and loop operators associated with the topological symmetries for the Ising conformal field theory in two space-time dimensions are realized on IBM's Kingston simulator. The relevant states are created on the quantum device using a hybrid quantum-classical algorithm. The latter is based on a variation of the quantum approximate optimization algorithm ansatz combined with the quantum natural gradient optimization method. Signatures of the topological symmetry are captured by measuring correlation functions of different qubit operators with results obtained from the quantum device in reasonable agreement with those obtained from classical computations. The current work demonstrates the viability of noisy quantum simulators as platforms for investigating low-dimensional quantum field theories with direct access to observables that are often difficult to probe in conventional condensed matter experiments.

We study the most elementary family of cellular automata defined over an arbitrary group universe $G$ and an alphabet $A$: the lazy cellular automata, which act as the identity on configurations in $A^G$, except when they read a unique active transition $p \in A^S$, in which case they write a fixed symbol $a \in A$. As expected, the dynamical behavior of lazy cellular automata is relatively simple, yet subtle questions arise since they completely depend on the choice of $p$ and $a$. In this paper, we investigate the order of a lazy cellular automaton $\tau : A^G \to A^G$, defined as the cardinality of the set $\{ \tau^k : k \in \mathbb{N} \}$. In particular, we establish a general upper bound for the order of $\tau$ in terms of $p$ and $a$, and we prove that this bound is attained when $p$ is a quasi-constant pattern.

Large Language Models (LLMs) are increasingly deployed as autonomous agents, yet their practical utility is fundamentally constrained by a limited context window and state desynchronization resulting from the LLMs' stateless nature and inefficient context management. These limitations lead to unreliable output, unpredictable behavior, and inefficient resource usage, particularly when interacting with large, structured, and sensitive knowledge systems such as codebases and documents. To address these challenges, we introduce the Gatekeeper Protocol, a novel, domain-agnostic framework that governs agent-system interactions. Our protocol mandates that the agent first operate and reason on a minimalist, low-fidelity "latent state" representation of the system to strategically request high-fidelity context on demand. All interactions are mediated through a unified JSON format that serves as a declarative, state-synchronized protocol, ensuring the agent's model of the system remains verifiably grounded in the system's reality. We demonstrate the efficacy of this protocol with Sage, a reference implementation of the Gatekeeper Protocol for software development. Our results show that this approach significantly increases agent reliability, improves computational efficiency by minimizing token consumption, and enables scalable interaction with complex systems, creating a foundational methodology for building more robust, predictable, and grounded AI agents for any structured knowledge domain.

We show that if $X$ is a Stein space and, if $\Omega \subset X$ is exhaustable by a sequence $\Omega_1 \subset \Omega_2 \subset \ldots \subset \Omega_n \subset \ldots$ of open Stein subsets of $X$, then $\Omega$ is Stein. This generalizes a well-known result of Behnke and Stein which is obtained for $X=\mathbb{C}^n$ and solves the union problem, one of the most classical questions in Complex Analytic Geometry. When $X$ has dimension 2, we prove that the same result follows if we assume only that $\Omega \subset \subset X$ is a domain of holomorphy in a Stein normal space. It is known, however, that if $X$ is an arbitrary complex space which is exhaustable by an increasing sequence of open Stein subsets $X_1 \subset X_2 \subset \cdots \subset X_n \subset \cdots$, it does not follow in general that $X$ is holomorphically-convex or holomorphically-separate (even if $X$ has no singularities). One can even obtain 2-dimensional complex manifolds on which all holomorphic functions are constant.

The $SL_n$ spider gives a diagrammatic way to encode the representation category of the quantum group $U_q(sl_n)$. The aim of this paper is to define a new spider that contains the $SL_n$ spider. The new spider is defined by generators and relations, according to fairly simple rules that start with combinatorial data coming from the root system of $SL_n$.

The Robbins-Monro algorithm is a recursive, simulation-based stochastic procedure to approximate the zeros of a function that can be written as an expectation. It is known that under some technical assumptions, a Gaussian convergence can be established for the procedure. Here, we are interested in the local limit theorem, that is, quantifying this convergence on the density of the involved objects. The analysis relies on a parametrix technique for Markov chains converging to diffusions, where the drift is unbounded.

The method of 'coupling from the past' permits exact sampling from the invariant distribution of a Markov chain on a finite state space. The coupling is successful whenever the stochastic dynamics are such that there is coalescence of all trajectories. The issue of the coalescence or non-coalescence of trajectories of a finite state space Markov chain is investigated in this note. The notion of the 'coalescence number' $k(\mu)$ of a Markovian coupling $\mu$ is introduced, and results are presented concerning the set $K(P)$ of coalescence numbers of couplings corresponding to a given transition matrix $P$. Note: This is a revision of the original published version, in which part of Theorem 6 has been removed. A correction may be found in Thm 5.3 of arXiv:2510.13572.

We show the denominator formulas for the normalized $R$-matrix involving two arbitrary Kirillov--Reshetikhin (KR) modules $W^{(k)}_{m,a}$ and $W^{(l)}_{p,b}$ in all nonexceptional affine types, $D_4^{(3)}$, and $G_2^{(1)}$. To achieve our goal, we prove the existence of homomorphisms, which can be understood as generalization of Dorey rule to KR modules. We also conjecture a uniform denominator formulas for all simply-laced types; in particular, type $E_n^{(1)}$. With the denominator formulas, we determine the simplicity of tensor product of KR modules and degrees of poles of normalized $R$-matrices between two KR modules completely in nonexceptional affine types, $D_4^{(3)}$, and $G_2^{(1)}$. As an application, we prove that the certain sets of KR modules for the untwisted affine types, suggested by Hernandez and Leclerc as clusters, form strongly commuting families, which implies that all cluster monomials in the clusters are real simple modules.

A research for a general expression of the triplets of integer sided triangles with a $120^{\circ}$ angle, in parallel with the case of a $60^{\circ}$ angle.

Faltings proved that there are finitely many abelian varieties of genus $g$ over a number field $K$, with good reduction outside a finite set of primes $S$. Fixing one of these abelian varieties $A$, we prove that there are finitely many smooth hypersurfaces in $A$, with good reduction outside $S$, representing a given ample class in the Néron-Severi group of $A$, up to translation, as long as the dimension of $A$ is at least $4$. Our approach builds on the approach of arXiv:1807.02721 which studies $p$-adic variations of Hodge structure to turn finiteness results for $p$-adic Galois representations into geometric finiteness statements. A key new ingredient is an approach to proving big monodromy for the variations of Hodge structure arising from the middle cohomology of these hypersurfaces using the Tannakian theory of sheaf convolution on abelian varieties.

We investigate three types of Internal Diffusion Limited Aggregation (IDLA) models. These models are based on simple random walks on $\mathbf{Z}^2$ with infinitely many sources that are the points of the vertical axis $I(\infty)=\{0\}\times\mathbf{Z}$. Various properties are provided, such as stationarity, mixing, stabilization and shape theorems. Our results allow us to define a new directed (w.r.t. the horizontal direction) random forest spanning $\mathbf{Z}^2$, based on an IDLA protocol, which is invariant in distribution w.r.t. vertical translations.

Assuming the conformal field theory correlation functions defined on complex curves satisfy the recursion formulas, we express the corresponding cohomology theory via the generalizations of holomorphic connections. The cohomology is explicitly found in terms of higher genus counterparts of elliptic functions as analytic continuations of solutions for functional equations. Explicit examples associated to correlation functions on various genera are provided.

We define a natural notion of higher order stability and show that subsets of $\mathbb{F}_p^n$ that are tame in this sense can be approximately described by a union of low-complexity quadratic varieties, up to linear error. This generalizes the arithmetic regularity lemma for stable subsets of $\mathbb{F}_p^n$, proved in earlier work of the authors, to the realm of higher-order Fourier analysis. This result is strictly stronger than the structure theorem for sets of bounded $\mathrm{VC}_2$-dimension, first proved by the authors in earlier versions of this paper and now available as a separate manuscript arXiv:2510.12867. Taken together, these results provide group theoretic analogues of results obtained for 3-uniform hypergraphs in arXiv:2111.01737.

Let $X$ be a Calabi--Yau manifold and $Q\subset X$ a closed connected embedded special Lagrangian; closed Lagrangians mean compact Lagrangian submanifolds without boundary. We prove that if the fundamental group $\pi_1Q$ is abelian then there exists a Weinstein neighbourhood of $Q\subset X$ in which every closed irreducibly immersed special Lagrangian with unobstructed Floer cohomology is $C^1$ close to $Q.$ We prove also that if $\pi_1Q$ is virtually solvable then for every positive integer $R$ there exists a Weinstein neighbourhood of $Q\subset X$ in which every closed irreducibly immersed special Lagrangian of degree $\le R$ and with unobstructed Floer cohomology is unbranched; that is, the projection $L\to Q$ is a covering map. We prove a stronger statement when $\pi_1Q$ is finite and a weaker statement when $\pi_1Q$ has no non-abelian free subgroups. The $\pi_1Q$ conditions, the Floer cohomology condition and the special Lagrangian condition are all essential as we show by counterexamples.

We consider a general class of two-stage distributionally robust optimization (DRO) problems where the ambiguity set is constrained by fixed marginal probability laws that are not necessarily discrete. We derive primal and dual formulations of this class of problems and subsequently develop a numerical algorithm for computing approximate optimizers as well as approximate worst-case probability measures. Moreover, our algorithm computes both an upper bound and a lower bound for the optimal value of the problem, where the difference between the computed bounds provides a direct sub-optimality estimate of the computed solution. Most importantly, the sub-optimality can be controlled to be arbitrarily close to 0 by appropriately choosing the inputs of the algorithm. To demonstrate the effectiveness of the proposed algorithm, we apply it to three prominent instances of two-stage DRO problems in task scheduling, multi-product assembly, and supply chain network design with edge failure. The ambiguity sets in these problem instances involve a large number of continuous or discrete marginals. The numerical results showcase that the proposed algorithm computes high-quality robust decisions along with non-conservative sub-optimality estimates.

We study the positivity and causality axioms for Markov categories as properties of dilations and information flow in Markov categories, and in variations thereof for arbitrary semicartesian monoidal categories. These help us show that being a positive Markov category is merely an additional property of a symmetric monoidal category (rather than extra structure). We also characterize the positivity of representable Markov categories and prove that causality implies positivity, but not conversely. Finally, we note that positivity fails for quasi-Borel spaces and interpret this failure as a privacy property of probabilistic name generation.

The stack number of a directed acyclic graph $G$ is the minimum $k$ for which there is a topological ordering of $G$ and a $k$-coloring of the edges such that no two edges of the same color cross, i.e., have alternating endpoints along the topological ordering. We prove that the stack number of directed acyclic outerplanar graphs is bounded by a constant, which gives a positive answer to a conjecture by Heath, Pemmaraju and Trenk [SIAM J. Computing, 1999]. As an immediate consequence, this shows that all upward outerplanar graphs have constant stack number, answering a question by Bhore et al. [Eur. J. Comb., 2023] and thereby making significant progress towards the problem for general upward planar graphs originating from Nowakowski and Parker [Order, 1989]. As our main tool we develop the novel technique of directed $H$-partitions, which might be of independent interest. We complement the bounded stack number for directed acyclic outerplanar graphs by constructing a family of directed acyclic 2-trees that have unbounded stack number, thereby refuting a conjecture by Nöllenburg and Pupyrev [GD 2023].

In this paper, we study the moduli space of unipotent Weil-Deligne representations valued in a split reductive group $G$ and characterise which irreducible components are smooth. We apply the smoothness results proved to show that a certain space of ordinary automorphic forms is a locally generically free module over the corresponding global deformation ring.

The Principle of Maximum Entropy is a rigorous technique for estimating an unknown distribution given partial information while simultaneously minimizing bias. However, an important requirement for applying the principle is that the available information be provided error-free (Jaynes 1982). We relax this requirement using a memoryless communication channel as a framework to derive a new, more general principle. We show our new principle provides an upper bound on the entropy of the unknown distribution and the amount of information lost due to the use of a given communications channel is unknown unless the unknown distribution's entropy is also known. Using our new principle we provide a new interpretation of the classic principle and experimentally show its performance relative to the classic principle and other generally applicable solutions. Finally, we present a simple algorithm for solving our new principle and an approximation useful when samples are limited.

A family $\mathcal{F}$ of sets satisfies the $(p,q)$-property if among every $p$ members of $\mathcal{F}$, some $q$ can be pierced by a single point. The celebrated $(p,q)$-theorem of Alon and Kleitman asserts that for any $p \geq q \geq d+1$, any family $\mathcal{F}$ of compact convex sets in $\mathbb{R}^d$ that satisfies the $(p,q)$-property can be pierced by a finite number $c(p,q,d)$ of points. A similar theorem with respect to piercing by $(d-1)$-dimensional flats, called $(d-1)$-transversals, was obtained by Alon and Kalai. In this paper we prove the following result, which can be viewed as an $(\aleph_0,k+2)$-theorem with respect to $k$-transversals: Let $\mathcal{F}$ be an infinite family of closed balls in $\mathbb{R}^d$, and let $0 \leq k < d$. If among every $\aleph_0$ elements of $\mathcal{F}$, some $k+2$ can be pierced by a $k$-dimensional flat, then $\mathcal{F}$ can be pierced by a finite number of $k$-dimensional flats. We derive this result as a corollary of a more general result which proves the same assertion for families of not necessarily convex objects called \emph{near-balls}, to be defined below. This is the first $(p,q)$-theorem in which the assumption is weakened to an $(\infty,\cdot)$ assumption. Our proofs combine geometric and topological tools.

We give an historical survey of some of the original basic algebraic and combinatorial results on Temperley-Lieb algebras, with a focus on certain results that have become folklore.

We study several notions of dimension for (pre-)triangulated categories naturally arising from topology and symplectic geometry. We prove new bounds on these dimensions and raise several questions for further investigation. For instance, we relate the Rouquier dimension of the wrapped Fukaya category of either the cotangent bundle of a smooth manifold $M$ or more generally a Weinstein domain $X$ to quantities of geometric interest. These quantities include the minimum number of critical values of a Morse function on $M$, the Lusternik-Schnirelmann category of $M$, the number of distinct action values of a Hamiltonian diffeomorphism of $X$, and the smallest $n$ such that $X$ admits a Weinstein embedding into $\mathbb{R}^{2n+1}$. Along the way, we introduce a notion of the Lusternik-Schnirelmann category for dg-categories and construct exact Lagrangian cobordisms for restriction to a Liouville subdomain.

We build and study a recursive algorithm based on the occupation measure of an Euler scheme with decreasing step for the numerical approximation of the quasistationary distribution (QSD) of an elliptic diffusion in a bounded domain. We prove the almost sure convergence of the procedure for a family of redistributions and show that we can also recover the approximation of the rate of survival and the convergence in distribution of the algorithm. This last point follows from some new bounds on the weak error related to diffusion dynamics with renewal.

This study analyzes the derivative-free loss method to solve a certain class of elliptic PDEs and fluid problems using neural networks. The approach leverages the Feynman-Kac formulation, incorporating stochastic walkers and their averaged values. We investigate how the time interval associated with the Feynman-Kac representation and the walker size influence computational efficiency, trainability, and sampling errors. Our analysis shows that the training loss bias scales proportionally with the time interval and the spatial gradient of the neural network, while being inversely proportional to the walker size. Moreover, we demonstrate that the time interval must be sufficiently long to enable effective training. These results indicate that the walker size can be chosen as small as possible, provided it satisfies the optimal lower bound determined by the time interval. Finally, we present numerical experiments that support our theoretical findings.

A $k$-uniform hypergraph (or $k$-graph) $H = (V, E)$ is $k$-partite if $V$ can be partitioned into $k$ sets $V_1, \ldots, V_k$ such that each edge in $E$ contains precisely one vertex from each $V_i$. In this note, we consider list colorings for such hypergraphs. We show that for any $\varepsilon > 0$ if each vertex $v \in V(H)$ is assigned a list of size $|L(v)| \geq \left((k-1+\varepsilon)\Delta/\log \Delta\right)^{1/(k-1)}$, then $H$ admits a proper $L$-coloring, provided $\Delta$ is sufficiently large. Up to a constant factor, this matches the bound on the chromatic number of simple $k$-graphs shown by Frieze and Mubayi, and that on the list chromatic number of triangle free $k$-graphs shown by Li and Postle. Our results hold in the more general setting of ``color-degree'' as has been considered for graphs. Furthermore, we establish a number of asymmetric statements matching results of Alon, Cambie, and Kang for bipartite graphs.

We develop the stochastic two-scale convergence method in the framework of Orlicz-Sobolev spaces, in order to deal with the homogenization of coupled stochastic-periodic problems in such spaces. One fundamental in this topic is the extension of compactness results for this method to the Orlicz setting. For the application, we show that the sequence of minimizers of a class of highly oscillatory minimizations problems involving integral functionals with convex and nonstandard growth integrands, converges to the minimizer of a homogenized problem.

We study representations of finite groups on Stanley--Reisner rings of simplicial complexes and on lattice points in lattice polytopes. The framework of translative group actions allows us to use the theory of proper colorings of simplicial complexes without requiring an explicit coloring to be given. We prove that the equivariant Hilbert series of a Cohen--Macaulay simplicial complex under a translative group action admits a rational expression whose numerator is a positive integer combination of irreducible characters. This implies an analogous rational expression for the equivariant Ehrhart series of a lattice polytope with a unimodular triangulation that is invariant under a translative group action. As an application, we study the equivariant Ehrhart series of alcoved polytopes in the sense of Lam and Postnikov and derive explicit results in the case of order polytopes and of Lipschitz poset polytopes.

Given two measures $\mu$, $\nu$ on Rd that satisfy Carleman's condition, we provide a numerical scheme to approximate as closely as desired the total variation distance between $\mu$ and $\nu$. It consists of solving a sequence (hierarchy) of convex relaxations whose associated sequence of optimal values converges to the total variation distance, an additional illustration of the versatility of the Moment-SOS hierarchy. Indeed each relaxation in the hierarchy is a semidefinite program whose size increases with the number of involved moments. It has an optimal solution which is a couple of degree-2n pseudo-moments which converge, as n grows, to moments of the Hahn-Jordan decomposition of $\mu$-$\nu$.

We prove tight probabilistic bounds for the shortest vectors in module lattices over number fields using the results of arXiv:2308.15275. Moreover, establishing asymptotic formulae for counts of fixed rank matrices with algebraic integer entries and bounded Euclidean length, we prove an approximate Rogers integral formula for discrete sets of module lattices obtained from lifts of algebraic codes. This in turn implies that the moment estimates of arXiv:2308.15275 as well as the aforementioned bounds on the shortest vector also carry through for large enough discrete sets of module lattices.

While iterative matrix inversion methods excel in computational efficiency, memory optimization, and support for parallel and distributed computing when managing large matrices, their limitations are also evident in multiple-input multiple-output (MIMO) fading channels. These methods encounter challenges related to slow convergence and diminished accuracy, especially in ill-conditioned scenarios, hindering their application in future MIMO networks such as extra-large aperture array. To address these challenges, this paper proposes a novel matrix regularization method termed symmetric rank-$1$ regularization (SR-$1$R). The proposed method functions by augmenting the channel matrix with a symmetric rank-$1$ matrix, with the primary goal of minimizing the condition number of the resultant regularized matrix. This significantly improves the matrix condition, enabling fast and accurate iterative inversion of the regularized matrix. Then, the inverse of the original channel matrix is obtained by applying the Sherman-Morrison transform on the outcome of iterative inversions. Our eigenvalue analysis unveils the best channel condition that can be achieved by an optimized SR-$1$R matrix. Moreover, a power iteration-assisted (PIA) approach is proposed to find the optimum SR-$1$R matrix without need of eigenvalue decomposition. The proposed approach exhibits logarithmic algorithm-depth in parallel computing for MIMO precoding. Finally, computer simulations demonstrate that SR-$1$R has the potential to reduce the required iteration by up to $35\%$ while achieving the performance of regularized zero-forcing.

We prove that compact Calabi--Yau varieties with certain isolated singularities are projective. In dimension 3 we do this by analysis, supposing given conifold metrics. In higher dimensions it follows more readily from Ohsawa's degenerate spectral sequence.

In this paper, we propose a novel gradient-free and dimensionality-independent sampler, the Geometric Optics Approximation Sampling (GOAS), based on a near-field reflector system. The key idea involves constructing a reflecting surface that redirects rays from a source with a prescribed simple distribution toward a target domain, achieving the desired target measure. Once this surface is constructed, an arbitrary number of independent, uncorrelated samples can be drawn by re-simulating (ray-tracing) the reflector system, i.e., push-forward samples from the source distribution under a reflecting map. To compute the reflecting surface, we employ an enhanced supporting ellipsoid method for the near-field reflector problem. This approach does not require gradient information of the target density and discretizes the target measure using either a low-discrepancy or random sequence, ensuring dimensionality independence. Since the resulting surface is non-smooth (being a union of ellipsoidal sheets) but continuous, we apply a softmin smoothing technique to enable sampling. Theoretically, we define the geometric optics approximation measure as the push-forward of the source measure through the reflecting map. We prove that this measure is well-defined and stable with respect to perturbations of the target domain, ensuring robustness in sampling. Additionally, we derive error bounds between the numerical geometric optics approximation measure and the target measure under the Hellinger metric. Our numerical experiments validate the theoretical claims of GOAS, demonstrate its superior performance compared to MCMC for complex distributions, and confirm its practical effectiveness and broad applicability in solving Bayesian inverse problems.

Kontsevich's formula for rational plane curves is a recursive relation for the number $N_d$ of degree $d$ rational curves in $\mathbb{P}^2$ passing through $3d-1$ general points. We provide two proofs of this recursion: the first more direct and combinatoric, the second more abstract. In order to achieve this, we introduce several moduli spaces, such as the Deligne-Mumford-Knudsen spaces and the Kontsevich spaces, and exploit their properties. In particular, the boundary structure of these spaces gives rise to certain fundamental relations crucial to both proofs. For the second proof, we reconsider the objects in question from the cohomological viewpoint and generalize the numbers $N_d$ to Gromov-Witten invariants. We introduce quantum cohomology and deduce Kontsevich's formula from the associativity of the quantum product. We also adapt these steps to the case of curves in $\mathbb{P}^1\times\mathbb{P}^1$, whose bidegrees lead to slightly more complicated but analogous results.

We study the existence and small scale behaviour of almost splitting maps along a Ricci flow satisfying Type I curvature bounds. These are special solutions of the heat equation that serve as parabolic analogues of harmonic almost splitting maps, which have proven to be an indespensable tool in the study of the structure of the singular set of non-collapsed Ricci limit spaces. In this paper, motivated by the recent work of Cheeger-Jiang-Naber in the Ricci limit setting, we construct sharp splitting maps on Ricci flows that are almost selfsimilar, and then investigate their small scale behaviour. We show that, modulo linear transformations, an almost splitting map at a large scale remains a splitting map even at smaller scales, provided that the Ricci flow remains sufficiently self-similar. Allowing these linear transformations means that a priori an almost splitting map might degenerate at small scales. However, we show that under an additional summability hypothesis such degeneration doesn't occur.

Let $A$ be an abelian variety defined over a global function field $F$, and let $p$ be a prime distinct from the characteristic of $F$. Let $F_\infty$ be a $p$-adic Lie extension of $F$ that contains the cyclotomic $\mathbb{Z}_p$-extension $F^{\mathrm{cyc}}$ of $F$. In this paper, we investigate the structure of the $p$-primary Selmer group $\mathrm{Sel}(A/F_\infty)$ of $A$ over $F_\infty$. We prove the $\mathfrak{M}_H(G)$-conjecture for $A/F_\infty$. Furthermore, we show that both the $\mu$-invariant of the Pontryagin dual of the Selmer group $\mathrm{Sel}(A/F^\mathrm{cyc})$ and the generalised $\mu$-invariant of the Pontryagin dual of the Selmer group $\mathrm{Sel}(A/F_\infty)$ are zero, therby proving Mazur's conjecture for $A/F$. We then relate the order of vanishing of the characteristic elements, evaluated at Artin representations, to the corank of the Selmer group of the corresponding twist of $A$ over the base field $F$. Assuming the finiteness of the Tate-Shafarevich group, we establish that this corank equals the order of vanishing of the $L$-function of $A/F$ at $s=1$. Finally, we extend a theorem of Sechi - originally proved for elliptic curves without complex multiplication - to abelian varieties over global function fields. This is achieved by adapting the notion of generalised Euler characteristic, introduced by Zerbes for elliptic curves over number fields. This new invariant allows us, via Akashi series, to relate the generalised Euler characteristic of $\mathrm{Sel}(A/F_\infty)$ to the Euler characteristic of $\mathrm{Sel}(A/F^{\mathrm{cyc}})$.

We look for travelling wave fields $$ E(x,y,z,t)= U(x,y) \cos(kz+\omega t)+ \widetilde U(x,y)\sin(kz+\omega t),\quad (x,y,z)\in\mathbb{R}^3,\, t\in\mathbb{R}, $$ satisfying Maxwell's equations in a nonlinear and cylindrically symmetric medium. We obtain a sequence of solutions with diverging energy that is different from that obtained by McLeod, Stuart, and Troy. In addition, we consider a more general nonlinearity, controlled by an \textit{N}-function.

Social welfare orders seek to combine the disparate preferences of an infinite sequence of generations into a single, societal preference order in some reasonably-equitable way. In [2] Dubey and Laguzzi study a type of social welfare order which they call SEA, for strongly equitable and (finitely) anonymous. They prove that the existence of a SEA order implies the existence of a set of reals which does not have the Baire property, and observe that a nonprincipal ultrafilter over $\mathbb{N}$ can be used to construct a SEA order. Questions arising in their work include whether the existence of a SEA order implies the existence of either a set of real numbers which is not Lebesgue-measurable or of a nonprincipal ultrafilter over $\mathbb{N}$. We answer both these questions, the solution to the second using the techniques of geometric set theory as set out by Larson and Zapletal in [11]. The outcome is that the existence of a SEA order does imply the existence of a set of reals which is not Lebesgue-measurable, and does not imply the existence of a nonprincipal ultrafilter on $\mathbb{N}$.

We consider a control problem for the nonlinear stochastic Fokker--Planck equation. This equation describes the evolution of the distribution of nonlocally interacting particles affected by a common source of noise. The system is directed by a controller that acts on the drift term with the goal of minimising a cost functional. We establish the well-posedness of the state equation, prove the existence of optimal controls, and formulate a stochastic maximum principle (SMP) that provides necessary and sufficient optimality conditions for the control problem. The adjoint process arising in the SMP is characterised by a nonlocal (semi)linear backward SPDE for which we study existence and uniqueness. We also rigorously connect the control problem for the nonlinear stochastic Fokker--Planck equation to the control of the corresponding McKean--Vlasov SDE that describes the motion of a representative particle. Our work extends existing results for the control of the Fokker--Planck equation to nonlinear and stochastic dynamics. In particular, the sufficient SMP, which we obtain by exploiting the special structure of the Fokker--Planck equation, seems to be novel even in the linear deterministic setting. We illustrate our results with an application to a model of government interventions in financial systems, supplemented by numerical illustrations.

Let $M$ be a holomorphically symplectic manifold, equipped with a Lagrangian fibration $\pi:\; M \to X$. A degenerate twistor deformation (sometimes also called ``a Tate-Shafarevich twist'') is a family of holomorphically symplectic structures on $M$ parametrized by $H^{1,1}(X)$. All members of this family are equipped with a holomorphic Lagrangian projection to $X$, and their fibers are isomorphic to the fibers of $\pi$. Assume that $M$ is a compact hyperkahler manifold of maximal holonomy, and the general fiber of the Lagrangian projection $\pi$ is primitive (that is, not divisible) in integer homology. We also assume that $\pi$ has reduced fibers in codimension 1. Then $M$ has a degenerate twistor deformation $M'$ such that the Lagrangian projection $\pi:\; M' \to X$ admits a meromorphic section.

Testing by betting has been a cornerstone of the game-theoretic statistics literature. In this framework, a betting score (or more generally an e-process), as opposed to a traditional p-value, is used to quantify the evidence against a null hypothesis: the higher the betting score, the more money one has made betting against the null, and thus the larger the evidence that the null is false. A key ingredient assumed throughout past works is that one cannot bet more money than one currently has. In this paper, we ask what happens if the bettor is allowed to borrow money after going bankrupt, allowing further financial flexibility in this game of hypothesis testing. We propose various definitions of (adjusted) evidence relative to the wealth borrowed, indebted, and accumulated. We also ask what happens if the bettor can "bargain", in order to obtain odds bettor than specified by the null hypothesis. The adjustment of wealth in order to serve as evidence appeals to the characterization of arbitrage, interest rates, and numéraire-adjusted pricing in this setting.

Partial differential equations (PDEs) are widely used for modeling various physical phenomena. These equations often depend on certain parameters, necessitating either the identification of optimal parameters or the solution of the equations over multiple parameters. Performing an exhaustive search over the parameter space requires solving the PDE multiple times, which is generally impractical. To address this challenge, reduced order models (ROMs) are built using a set of precomputed solutions (snapshots) corresponding to different parameter values. Recently, Deep Learning ROMs (DL-ROMs) have been introduced as a new method to obtain ROM, offering improved flexibility and performance. In many cases, the domain on which the PDE is defined also varies. Capturing this variation is important for building accurate ROMs but is often difficult, especially when the domain has a complex structure or changes topology. In this paper, we propose a Deep-ROM framework that can automatically extract useful domain parametrization and incorporate it into the model. Unlike traditional domain parameterization methods, our approach does not require user-defined control points and can effectively handle domains with varying numbers of components. It can also learn from domain data even when no mesh is available. Using deep autoencoders, our approach reduces the dimensionality of both the PDE solution and the domain representation, making it possible to approximate solutions efficiently across a wide range of domain shapes and parameter values. We demonstrate that our approach produces parametrizations that yield solution accuracy comparable to models using exact parameters. Importantly, our model remains stable under moderate geometric variations in the domain, such as boundary deformations and noise - scenarios where traditional ROMs often require remeshing or manual adjustment.

We study homological mirror symmetry for $(\mathbb{P}^2, \Omega)$ viewed as an object of birational geometry, with $\Omega$ the standard meromorphic volume form. First, we construct universal objects on the two sides of mirror symmetry, focusing on the exact symplectic setting: a smooth complex scheme $U_\mathrm{univ}$ and a Weinstein manifold $M_\mathrm{univ}$, both of infinite type; and we prove homological mirror symmetry for them. Second, we consider autoequivalences. We prove that automorphisms of $U_\mathrm{univ}$ are given by a natural discrete subgroup of $\operatorname{Bir} (\mathbb{P}^2, \pm \Omega)$; and that all of these automorphisms are mirror to symplectomorphisms of $M_\mathrm{univ}$. We conclude with some applications.

In this short paper, we aim at giving a more conceptual and simpler proof of Rumely's moduli theoretic characterization of type II minimal locus of the resultant function $\operatorname{ordRes}_\phi$ on the Berkovich hyperbolic space for a rational function $\phi$ on $\mathbb{P}^1$ defined over an algebraically closed and complete field that is equipped with a non-trivial and non-archimedean absolute value, and also aim at giving a much simpler and more natural proof of a degenerating limit theorem, in an improved form after DeMarco--Faber, for the family of the unique maximal entropy measures on $\mathbb{P}^1(\mathbb{C})$ associated to a meromorphic family of complex rational functions. We introduce the intrinsic reduction of a non-archimedean rational function $\phi$ at each point in the Berkovich projective line and its directionwise intrinsic depths, which are suitable notions for the above aims and defined in terms of the tree and analytic structures of the Berkovich projective line. Then we establish two theorems in non-archimedean dynamics, both of which play key roles in the above aims.

Deep neural network (DNN)-based algorithms are emerging as an important tool for many physical and MAC layer functions in future wireless communication systems, including for large multi-antenna channels. However, training such models typically requires a large dataset of high-dimensional channel measurements, which are very difficult and expensive to obtain. This paper introduces a novel method for generating synthetic wireless channel data using diffusion-based models to produce user-specific channels that accurately reflect real-world wireless environments. Our approach employs a conditional denoising diffusion implicit model (cDDIM) framework, effectively capturing the relationship between user location and multi-antenna channel characteristics. We generate synthetic high fidelity channel samples using user positions as conditional inputs, creating larger augmented datasets to overcome measurement scarcity. The utility of this method is demonstrated through its efficacy in training various downstream tasks such as channel compression and beam alignment. Our diffusion-based augmentation approach achieves over a 1-2 dB gain in NMSE for channel compression, and an 11dB SNR boost in beamforming compared to prior methods, such as noise addition or the use of generative adversarial networks (GANs).

We discuss equivalent formulations of the control of conditional processes introduced by Lions. In this problem, a controlled diffusion process is killed once it hits the boundary of a given domain and the controller's reward is computed based on the conditional distribution given the process's survival. So far there is no clarity regarding the relationship between the open- and closed-loop formulation of this nonstandard control problem. We provide a short proof of their equivalence using measurable selection and mimicking arguments. In addition, we link the closed-loop formulation to Fleming--Viot dynamics of McKean--Vlasov type, where upon being killed the diffusion process is reinserted into the domain according to the current law of the process itself. This connection offers a new interpretation of the control problem and opens it up to applications that feature costs caused by reinsertion.

There is no known polynomial-time algorithm for graph isomorphism testing, but elementary combinatorial "refinement" algorithms seem to be very efficient in practice. Some philosophical justification is provided by a classical theorem of Babai, Erdős and Selkow: an extremely simple polynomial-time combinatorial algorithm (variously known as "naïve refinement", "naïve vertex classification", "colour refinement" or the "1-dimensional Weisfeiler-Leman algorithm") yields a so-called canonical labelling scheme for "almost all graphs". More precisely, for a typical outcome of a random graph $G(n,1/2)$, this simple combinatorial algorithm assigns labels to vertices in a way that easily permits isomorphism-testing against any other graph. We improve the Babai-Erdős-Selkow theorem in two directions. First, we consider randomly perturbed graphs, in accordance with the smoothed analysis philosophy of Spielman and Teng: for any graph $G$, naïve refinement becomes effective after a tiny random perturbation to $G$ (specifically, the addition and removal of $O(n\log n)$ random edges). Actually, with a twist on naïve refinement, we show that $O(n)$ random additions and removals suffice. These results significantly improve on previous work of Gaudio-Rácz-Sridhar, and are in certain senses best-possible. Second, we complete a long line of research on canonical labelling of random graphs: for any $p$ (possibly depending on $n$), we prove that a random graph $G(n,p)$ can typically be canonically labelled in polynomial time. This is most interesting in the extremely sparse regime where $p$ has order of magnitude $c/n$; denser regimes were previously handled by Bollobás, Czajka-Pandurangan, and Linial-Mosheiff. Our proof also provides a description of the automorphism group of a typical outcome of $G(n,p_n)$ (slightly correcting a prediction of Linial-Mosheiff).

A function from $\Bbb F_{2^n}$ to $\Bbb F_{2^n}$ is said to be {\em $k$th order sum-free} if the sum of its values over each $k$-dimensional $\Bbb F_2$-affine subspace of $\Bbb F_{2^n}$ is nonzero. This notion was recently introduced by C. Carlet as, among other things, a generalization of APN functions. At the center of this new topic is a conjecture about the sum-freedom of the multiplicative inverse function $f_{\text{\rm inv}}(x)=x^{-1}$ (with $0^{-1}$ defined to be $0$). It is known that $f_{\text{\rm inv}}$ is 2nd order (equivalently, $(n-2)$th order) sum-free if and only if $n$ is odd, and it is conjectured that for $3\le k\le n-3$, $f_{\text{\rm inv}}$ is never $k$th order sum-free. The conjecture has been confirmed for even $n$ but remains open for odd $n$. In the present paper, we show that the conjecture holds under each of the following conditions: (1) $n=13$; (2) $3\mid n$; (3) $5\mid n$; (4) the smallest prime divisor $l$ of $n$ satisfies $(l-1)(l+2)\le (n+1)/2$. We also determine the ``right'' $q$-ary generalization of the binary multiplicative inverse function $f_{\text{\rm inv}}$ in the context of sum-freedom. This $q$-ary generalization not only maintains most results for its binary version, but also exhibits some extraordinary phenomena that are not observed in the binary case.

In this paper, we generalize the minimum flow decomposition problem (MFD) to incorporate uncertain edge capacities and tackle it from the perspective of robust optimization. In the classical flow decomposition problem, a network flow is decomposed into a set of weighted paths from a fixed source node to a fixed sink node that precisely represents the flow distribution across all edges. MFD problems permeate multiple important applications, including reconstructing genomic sequences to representing the flow of goods or passengers in distribution networks. Inspired by these applications, we generalize the MFD to an inexact case with bounded flow values, provide a detailed analysis, and explore different variants that are solvable in polynomial time. Moreover, we introduce the concept of robust flow decomposition by incorporating uncertain bounds and applying different robustness concepts to handle the uncertainty. Finally, we present two different adjustably robust problem formulations and perform computational experiments illustrating the benefit of adjustability.

In Special Relativity, massless objects are characterized as either vacuum states or as radiation propagating at the speed of light. This distinction extends to General Relativity for asymptotically flat initial data sets (IDS) \((M^n, g, k)\), where vacuum is represented by slices of Minkowski space, and radiation is modeled by slices of \(pp\)-wave spacetimes. In contrast, we demonstrate that asymptotically hyperboloidal IDS with zero mass must embed isometrically into Minkowski space, with no possible IDS configurations modeling radiation in this setting. Our result holds under the most general assumptions. The proof relies on precise decay estimates for spinors on level sets of spacetime harmonic functions and works in all dimensions.

We investigate compatibility of gradings for an almost Koszul or Koszul algebra $R$ that is also the higher preprojective algebra $\Pi_{n+1}(A)$ of an $n$-hereditary algebra $A$. For an $n$-representation finite algebra $A$, we show that $A$ must be Koszul if $\Pi_{n+1}(A)$ can be endowed with an almost Koszul grading. For an acyclic basic $n$-representation infinite algebra $A$, we show that $A$ must be Koszul if $\Pi_{n+1}(A)$ can be endowed with a Koszul grading. From this we deduce that a higher preprojective grading of an (almost) Koszul algebra $R = \Pi_{n+1}(A)$ is, in both cases, isomorphic to a cut of the (almost) Koszul grading. Up to a further assumption on the tops of the degree $0$ subalgebras for the different gradings, we also show a similar result without the basic assumption in the $n$-representation infinite case. As an application, we show that $n$-APR tilting preserves the property of being Koszul for $n$-representation infinite algebras.

The monotone path polytope of a polytope $P$ encapsulates the combinatorial behavior of the shadow vertex rule (a pivot rule used in linear programming) on $P$. Computing monotone path polytopes is the entry door to the larger subject of fiber polytopes, for which explicitly computing examples remains a challenge. We first give a detailed presentation on how to construct monotone path polytopes. Monotone path polytopes of cubes and simplices have been known since the seminal article of Billera and Sturmfels. We extend these results to hypersimplices by linking this problem to the combinatorics of lattice paths. Indeed, we give a combinatorial model which describes the vertices of the monotone path polytope of the hypersimplex $\Delta(n, 2)$ (for any generic direction). With this model, we give a precise count of these vertices, and furthermore count the number of coherent monotone paths on $\Delta(n, 2)$ according to their lengths. We prove that some of the results obtained also hold for hypersimplices $\Delta(n, k)$ for $k\geq 2$.

In this paper we examine two ways of coding sequences in arithmetical theories. We investigate under what conditions they work. To be more precise, we study the creation of objects of a data-type that we call ur-strings, roughly sequences where the components are ordered but where we do not have an explicitly given projection function. First, we have a brief look at the beta-function which was already carefully studied by Emil Jeřábek. We study in detail our two target constructions. These constructions both employ theories of strings. The first is based on Smullyan coding and the second on the representation of binary strings in the special linear monoid of the non-negative part of discretely ordered commutative rings as introduced by Markov. We use the Markov coding to obtain an alternative proof that ${\sf PA}^{-}$ is sequential.

We study a random configuration of $N$ soliton solutions $\psi_N(x,t;\boldsymbol{\lambda})$ of the cubic focusing Nonlinear Schrödinger (fNLS) equation in one space dimension. The $N$ soliton solutions are parametrized by $2N$ complex numbers $(\boldsymbol{\lambda}, \boldsymbol{c})$ where $\boldsymbol{\lambda}\in\mathbb{C}_+^N$ are the eigenvalues of the Zakharov-Shabat linear operator, and $ \boldsymbol{c}\in\mathbb{C}^N\backslash \{0\}$ are the norming constants of the corresponding eigenfunctions. The randomness is obtained by choosing the complex eigenvalues to be i.i.d. random variables sampled from a probability distribution with compact support in the complex plane. The corresponding norming constants are interpolated by a smooth function of the eigenvalues. Then we consider the expectation of the random measure associated to this random spectral data. Such expectation uniquely identifies, via the Zakharov-Shabat inverse spectral problem, a solution $\psi_\infty(x,t)$ of the fNLS equation. This solution can be interpreted as a soliton gas solution. We prove a Law of Large Numbers and a Central Limit Theorem for the differences $\psi_N(x,t;\boldsymbol{\lambda})-\psi_\infty(x,t)$ and $|\psi_N(x,t;\boldsymbol{\lambda})|^2-|\psi_\infty(x,t)|^2$ when $(x,t)$ are in a compact set of $\mathbb R\times\mathbb R^+$; we additionally compute the correlation functions.

Generically, small deformations of cone manifold holonomy groups have wildly uncontrolled global geometry. We give a short concrete example showing that it is possible to deform complete hyperbolic metrics on a thickened genus $2$ surface to complete hyperbolic metrics on the genus two handlebody with a single unknotted cusp drilled out via cone manifolds of prescribed singular structure. In other words, there exists a method to construct smooth curves in the character variety of $ \pi_1(S_{2,0}) $ which join open sets parameterising discrete groups (quasi-conformal deformation spaces) through indiscrete groups where the indiscreteness arises in a very controlled, local, way: a cone angle increasing along a fixed singular locus.

10 is the smallest positive integer which is whether solitary or friendly is still an open question in mathematics. In this paper, we provide upper bounds for each of the prime divisors of a friend of 10. This paper is precisely a generalization of a recent paper [4] in which necessary upper bounds for the 2nd, 3rd, and 4th smallest prime divisors of a friend of 10 have been proved. Further, we establish better upper bounds for the 3rd, and 4th smallest prime divisors of a friend of 10 than the bounds given in [4].

A random planar quadrangulation process is introduced as an approximation for certain cellular automata in terms of random growth of rays from a given set of points. This model turns out to be a particular (rectangular) case of the well-known Gilbert tessellation, which originally models the growth of needle-shaped crystals from the initial random points with a Poisson distribution in a plane. From each point the lines grow on both sides of vertical and horizontal directions until they meet another line. This process results in a rectangular tessellation of the plane. The central and still open question is the distribution of the length of line segments in this tessellation. We derive exponential bounds for the tail of this distribution. The correlations between the segments are proved to decay exponentially with the distance between their initial points. Furthermore, the sign of the correlation is investigated for some instructive examples. In the case when the initial set of points is confined in a box $[0,N]^2$, it is proved that the average number of rays reaching the border of the box has a linear order in $N$.

As a foundation for optimization, convexity is useful beyond the classical settings of Euclidean and Hilbert space. The broader arena of nonpositively curved metric spaces, which includes manifolds like hyperbolic space, as well as metric trees and more general CAT(0) cubical complexes, supports primal tools like proximal operations for geodesically convex functions. However, the lack of linear structure in such spaces complicates dual constructions like subgradients. To address this hurdle, we introduce a new type of subgradient for functions on Hadamard spaces, based on Busemann functions. Our notion supports generalizations of classical stochastic and incremental subgradient methods, with guaranteed complexity bounds. We illustrate with subgradient algorithms for $p$-mean problems in general Hadamard spaces, in particular computing medians in BHV tree space.

In this paper, we investigate saddle-node to saddle separatrix--loops that we term SNICeroclinic bifurcations. They are generic codimension-two bifurcations involving a heteroclinic loop between one non-hyperbolic and one hyperbolic saddle. A particular codimension-three case is the non-central SNICeroclinic bifurcation. We unfold this bifurcation in the minimal dimension (planar) case where the non-hyperbolic point is assumed to undergo a saddle-node bifurcation. Applying the method of Poincaré return maps, we present a minimal set of perturbations that captures all qualitatively distinct behaviors near a non-central SNICeroclinic loop. Specifically, we study how variation of the three unfolding parameters leads to transitions from heteroclinic and homoclinic loops, saddle-node on an invariant circle (SNIC), and periodic orbits as well as equilibria. We show that although the bifurcation has been largely unexplored in applications, it can act as an organizing center for transitions between various types of saddle-node and saddle separatrix loops. It is also a generic route to oscillations that are both born and destroyed via global bifurcations, compared to the commonly observed scenarios involving local (Hopf) and in some cases global (homoclinic or SNIC) bifurcations.

We construct an expansion of a real closed field by a multiplicative subgroup adapting Poizat's theory of green points. Its theory is strongly dependent, and every open set definable in a model of this theory is semialgebraic. We prove that the real field with a dense family of logarithmic spirals, proposed by Zilber, satisfies our theory.

In this note we write down a proof of the following well known fact, in order to make the literature more transparent. Let $\mathfrak{g}$ be a simple Lie algebra, then for any smooth curve $C$, the bundle underlying any $\mathfrak{g}$-Oper depends only on the curve and it is induced by the canonical $\text{Aut}\, O$ bundle $\text{Aut}_C$ on $C$.

We address the classical problem of constructing confidence intervals (CIs) for the mean of a distribution, given \(N\) i.i.d. samples, such that the CI contains the true mean with probability at least \(1 - \delta\), where \(\delta \in (0,1)\). We characterize three distinct learning regimes based on the minimum achievable limiting width of any CI as the sample size \(N_{\delta} \to \infty\) and \(\delta \to 0\). In the first regime, where \(N_{\delta}\) grows slower than \(\log(1/\delta)\), the limiting width of any CI equals the width of the distribution's support, precluding meaningful inference. In the second regime, where \(N_{\delta}\) scales as \(\log(1/\delta)\), we precisely characterize the minimum limiting width, which depends on the scaling constant. In the third regime, where \(N_{\delta}\) grows faster than \(\log(1/\delta)\), complete learning is achievable, and the limiting width of the CI collapses to zero, converging to the true mean. We demonstrate that CIs derived from concentration inequalities based on Kullback--Leibler (KL) divergences achieve asymptotically optimal performance, attaining the minimum limiting width in both sufficient and complete learning regimes for distributions in two families: single-parameter exponential and bounded support. Additionally, these results extend to one-sided CIs, with the width notion adjusted appropriately. Finally, we generalize our findings to settings with random per-sample costs, motivated by practical applications such as stochastic simulators and cloud service selection. Instead of a fixed sample size, we consider a cost budget \(C_{\delta}\), identifying analogous learning regimes and characterizing the optimal CI construction policy.

We present a formalization of Borel determinacy in the Lean 4 theorem prover. The formalization includes a definition of Gale-Stewart games and a proof of Martin's theorem stating that Borel games are determined. The proof closely follows Martin's "A purely inductive proof of Borel determinacy".

In this work, we present a study combining two approaches in the context of solving PDEs: the continuous finite element method (FEM) and more recent techniques based on neural networks. In recent years, physics-informed neural networks (PINNs) have become particularly interesting for rapidly solving PDEs, especially in high dimensions. However, their lack of accuracy can be a significant drawback in this context, hence the interest in combining them with FEM, for which error estimates are already known. The complete pipeline proposed here consists in modifying the classical FEM approximation spaces by taking information from a prior, chosen as the prediction of a neural network. On the one hand, this combination improves and certifies the prediction of neural networks, to obtain a fast and accurate solution. On the other hand, error estimates are proven, showing that such strategies outperform classical ones by a factor that depends only on the quality of the prior. We validate our approach with numerical results performed on parametric problems with 1D, 2D and 3D geometries. These experiments demonstrate that to achieve a given accuracy, a coarser mesh can be used with our enriched FEM compared to the standard FEM, leading to reduced computational time, particularly for parametric problems.

We give several equivalent characterisations of the maximal pro-2 quotients of real projective groups. In particular, for pro-2 real projective groups we provide a presentation in terms of generators and relations, and a purely cohomological characterisation. As a consequence we explicitly reconstruct such groups from their mod 2 cohomology rings.

We introduce and study operadic categories with cardinalities in finite sets and establish conditions under which their associated theories of operads and algebras are equivalent to the standard framework introduced in 2015 by Batanin and Markl. Our approach is particularly natural in applications to the operadic category of graphs and the related category of modular operads and their clones.

In this paper, we present two stepsize strategies for the extended Golden Ratio primal-dual algorithm (E-GRPDA) designed to address structured convex optimization problems in finite-dimensional real Hilbert spaces. The first rule features a non-increasing primal stepsize that remains bounded below by a positive constant and is updated adaptively at each iteration, eliminating the need to compute the Lipschitz constant of the gradient of the function and the norm of the operator, without using backtracking. The second stepsize rule is adaptive, adjusting based on the local smoothness of the smooth component function and the norm of the operator involved. In other words, we present an adaptive version of the E-GRPDA algorithm. We prove that E-GRPDA achieves an ergodic sublinear convergence rate with both stepsize rules, based on the function-value residual and constraint violation rather than on the so-called primal-dual gap function. Additionally, we establish an R-linear convergence rate for E-GRPDA with the first stepsize rule, under standard assumptions and with appropriately chosen parameters. Through numerical experiments on various convex optimization problems, we demonstrate the effectiveness of our approaches and compare their performance to existing ones.

The Fundamental Theorem of Algebra can be thought of as a statement about the real numbers as a space, considered as an algebraic set over the real numbers as a field. This paper introduces what it means for an algebraic set or affine variety over a field to be fundamental, in a way that encompasses the Fundamental Theorem of Algebra as a special case. The related concept of local fundamentality is introduced and its behavior developed. On the algebraic side, the notions of locally, geometrically, and generically unit-additive rings are introduced, thus complementing unit-additivity as previously defined by the author and Jay Shapiro. A number of results are extended from the previous joint paper from unit-additivity to local unit-additivity. It is shown that an affine variety is (locally) fundamental if and only if its coordinate ring is (locally) unit-additive. To do so, a theorem is proved showing that there are many equivalent definitions of local unit-additivity. Illustrative examples are sprinkled throughout.

Reproducing Kernel Hilbert Space (RKHS) embedding of probability distributions has proved to be an effective approach, via MMD (maximum mean discrepancy), for nonparametric hypothesis testing problems involving distributions defined over general (non-Euclidean) domains. While a substantial amount of work has been done on this topic, only recently have minimax optimal two-sample tests been constructed that incorporate, unlike MMD, both the mean element and a regularized version of the covariance operator. However, as with most kernel algorithms, the optimal test scales cubically in the sample size, limiting its applicability. In this paper, we propose a spectral-regularized two-sample test based on random Fourier feature (RFF) approximation and investigate the trade-offs between statistical optimality and computational efficiency. We show the proposed test to be minimax optimal if the approximation order of RFF (which depends on the smoothness of the likelihood ratio and the decay rate of the eigenvalues of the integral operator) is sufficiently large. We develop a practically implementable permutation-based version of the proposed test with a data-adaptive strategy for selecting the regularization parameter. Finally, through numerical experiments on simulated and benchmark datasets, we demonstrate that the proposed RFF-based test is computationally efficient and performs almost similarly (with a small drop in power) to the exact test.

A recently developed upscaling technique, the multicontinuum homogenization method, has gained significant attention for its effectiveness in modeling complex multiscale systems. This method defines multiple continua based on distinct physical properties and solves a series of constrained cell problems to capture localized information for each continuum. However, solving all these cell problems on very fine grids at every macroscopic point is computationally expensive, which is a common limitation of most homogenization approaches for non-periodic problems. To address this challenge, we propose a hierarchical multicontinuum homogenization framework. The core idea is to define hierarchical macroscopic points and solve the constrained problems on grids of varying resolutions. We assume that the local solutions can be represented as a combination of a linear interpolation of local solutions from preceding levels and an additional correction term. This combination is substituted into the original constrained problems, and the correction term is resolved using finite element (FE) grids of varying sizes, depending on the level of the macropoint. By normalizing the computational cost of fully resolving the local problem to $\mathcal{O}(1)$, we establish that our approach incurs a cost of $\mathcal{O}(L \eta^{(1-L)d})$, highlighting substantial computational savings across hierarchical layers $L$, coarsening factor $\eta$, and spatial dimension $d$. Numerical experiments validate the effectiveness of the proposed method in media with slowly varying properties, underscoring its potential for efficient multiscale modeling.

We prove a necessary criterion for the (non-)existence of nontrivial solutions to the Dirac equation $D\psi=i A \cdot_{Cl} \psi$ on Riemannian manifolds that are either closed or of bounded geometry. This generalizes a result of Rupert Frank and Michael Loss on $\mathbb{R}^n$ where the criterion relates the $L^n$-norm of $A$ to the Sobolev constant on $\mathbb{R}^n$. On Riemannian manifolds the role of the Sobolev constant will be replaced by the Yamabe invariant. If $n$ is odd, we show that our criterion is sharp on $\mathbb{S}^n$.

Tomography is a central tool in medical applications, allowing doctors to investigate patients' interior features. The Radon transform (in two dimensions) is commonly used to model the measurement process in parallel-beam CT. Suitable discretization of the Radon transform and its adjoint (called the backprojection) is crucial. The most commonly used discretization approach combines what we refer to as the ray-driven Radon transform with what we refer to as the pixel-driven backprojection, as anecdotal reports describe these as showing the best approximation performance. However, there is little rigorous understanding of induced approximation errors. These methods involve three discretization parameters: the spatial-, detector-, and angular resolutions. Most commonly, balanced resolutions are used, i.e., the same (or similar) spatial- and detector resolutions are employed. We present an interpretation of ray- and pixel-driven discretizations as `convolutional methods', a special class of finite-rank operators. This allows for a structured analysis that can explain observed behavior. In particular, we prove convergence in the strong operator topology of the ray-driven Radon transform and the pixel-driven backprojection under balanced resolutions, thus theoretically justifying this approach. In particular, with high enough resolutions one can approximate the Radon transform arbitrarily well. Numerical experiments corroborate these theoretical findings.

Tournaments are competitions between a number of teams, the outcome of which determines the relative strength or rank of each team. In many cases, the strength of a team in the tournament is given by a score. Perhaps, the most striking mathematical result on the tournament is Moon's theorem, which provides a necessary and sufficient condition for a feasible score sequence via majorization. To give a probabilistic interpretation of Moon's result, Aldous and Kolesnik introduced the football model, the existence of which gives a short proof of Moon's theorem. However, the existence proof of Aldous and Kolesnik is nonconstructive, leading to the question of a ``canonical'' construction of the football model. The purpose of this paper is to provide explicit constructions of the football model with an additional stochastic ordering constraint, which can be formulated by martingale transport. Two solutions are given: one is by solving an entropy optimization problem via Sinkhorn's algorithm, and the other relies on the idea of shadow couplings. It turns out that both constructions yield the property of strong stochastic transitivity. The nontransitive situations of the football model are also considered.

A 2009 article of Allcock and Vaaler explored the $\mathbb Q$-vector space $\mathcal G := \overline{\mathbb Q}^\times/{\overline{\mathbb Q}^\times_{\mathrm{tors}}}$, showing how to represent it as part of a function space on the places of $\overline{\mathbb Q}$. We establish a representation theorem for the $\mathbb R$-vector space of $\mathbb Q$-linear maps from $\mathcal G$ to $\mathbb R$, enabling us to classify extensions to $\mathcal G$ of completely additive arithmetic functions. We further outline a strategy to construct $\mathbb Q$-linear maps from $\mathcal G$ to $\mathbb Q$, i.e., elements of the algebraic dual of $\mathcal G$. Our results make heavy use of Dirichlet's $S$-unit Theorem as well as a measure-like object called a consistent map, first introduced by the author in previous work.

We study the spectrum of limit models assuming the existence of a nicely behaved independence notion. Under reasonable assumptions, we show that all `long' limit models are isomorphic, and all `short' limit models are non-isomorphic. $\textbf{Theorem.}$ Let $\mathbf{K}$ be a $\aleph_0$-tame abstract elementary class stable in $\lambda \geq \operatorname{LS}(\mathbf{K})$ with amalgamation, joint embedding and no maximal models. Suppose there is an independence relation on the models of size $\lambda$ that satisfies uniqueness, extension, non-forking amalgamation, universal continuity, and $(\geq \kappa)$-local character in a minimal regular $\kappa < \lambda^+$. Suppose $\delta_1, \delta_2 < \lambda^+$ with $\operatorname{cf}(\delta_1) < \operatorname{cf}(\delta_2)$. Then for any $N_1, N_2, M \in \mathbf{K}_\lambda$ where $N_l$ is a $(\lambda, \delta_l)$-limit model over $M$ for $l = 1, 2$, \[N_1 \text{ is isomorphic to } N_2 \text{ over } M \iff \operatorname{cf}(\delta_1) \geq \kappa\] Both implications in the conclusion have improvements. High cofinality limits are isomorphic without the $\aleph_0$-tameness assumption and assuming the independence relation is defined only on high cofinality limit models. Low cofinality limits are non-isomorphic without assuming non-forking amalgamation. We show how our results can be used to study limit models in both abstract settings and in natural examples of abstract elementary classes.

Let $\Gamma$ be a graph with the adjacency matrix $A$. The transition matrix of $\Gamma$, denoted $H(t)$, is defined as $H(t) := \exp(-\textbf{i}tA)$, where $\textbf{i} := \sqrt{-1}$ and $t$ is a real variable. The graph $\Gamma$ is said to exhibit fractional revival (FR in short) between the vertices $a$ and $b$ if there exists a positive real number $t$ such that $H(t){\textbf{e}_{a}} = \alpha{\textbf{e}_{a}} + \beta{\textbf{e}_{b}}$, where $\alpha, \beta \in \mathbb{C}$ such that $\beta \neq 0$ and $|\alpha|^2 + |\beta|^2 = 1$. The graph $\Gamma$ is said to exhibit pretty good fractional revival (PGFR in short) between the vertices $a$ and $b$ if there exists a sequence of real numbers $\{t_k\}$ with $\lim_{k\to\infty} H(t_k){\textbf{e}_{a}} = \alpha{\textbf{e}_{a}} + \beta{\textbf{e}_{b}}$, where $\alpha, \beta \in \mathbb{C}$ such that $\beta \neq 0$ and $|\alpha|^2 + |\beta|^2 = 1$. In the definition of PGFR, if $\alpha=0$ then $\Gamma$ is said to exhibit pretty good state transfer (PGST in short) between $a$ and $b$. In this paper, we obtain some sufficient conditions for circulant graphs exhibiting PGFR. We also find some sufficient conditions for non-circulant abelian Cayley graphs exhibiting PGFR. From these sufficient conditions, we find infinite families of circulant graphs and non-circulant abelian Cayley graphs exhibiting PGFR that fail to exhibit FR and PGST. Finally, we obtain some necessary conditions for some families of circulant graphs exhibiting PGFR. Some of our results generalize the results of Chan et al. [Pretty good quantum fractional revival in paths and cycles. \textit {Algebr. Comb.} 4(6) (2021), 989-1004.] for cycles.

We determine the finite group $\mathcal S$ parametrizing a packet in the local Langlands correspondence for a Brylinski-Deligne covering group of an algebraic torus, under some assumption on ramification. Especially, this work generalizes Weissman's result on covering groups of tori that split over an unramified extension of the base field.

In a seminal work of Zhang and Sra, gradient descent methods for geodesically convex optimization were comprehensively studied. In particular, Zhang and Sra derived a comparison inequality that relates the iterative points in the optimization process. Since their seminal work, numerous follow-ups have studied different downstream usages of their comparison lemma. In this work, we introduce the concept of quasilinearization to optimization, presenting a novel framework for analyzing geodesically convex optimization. By leveraging this technique, we establish state-of-the-art convergence rates -- for both deterministic and stochastic settings -- under weaker assumptions than previously required. The technique of quasilinearization may prove valuable for other non-Euclidean optimization problems.

For a given symmetric association scheme $\mathcal{A}$ and its eigenspace $S_j$ there exists a mapping of vertices of $\mathcal{A}$ to unit vectors of $S_j$, known as the spherical representation of $\mathcal{A}$ in $S_j$, such that the inner products of these vectors only depend on the relation between the corresponding vertices; furthermore, these inner products only depend on the parameters of $\mathcal{A}$. We consider parameters of imprimitive association schemes listed as open cases in the list of parameters for quotient-polynomial graphs recently published by Herman and Maleki, and study embeddings of their substructures into some eigenspaces consistent with spherical representations of the putative association schemes. Using this, we obtain nonexistence for two parameter sets for $4$-class association schemes and one parameter sets for a $5$-class association scheme passing all previously known feasibility conditions, as well as uniqueness for two parameter sets for $5$-class association schemes.

The author proves that there are infinitely many primes $p$ such that $\| \alpha p - \beta \| < p^{-\frac{28}{87}}$, where $\alpha$ is an irrational number and $\beta$ is a real number. This sharpens a result of Jia (2000) and provides a new triple $(\gamma, \theta, \nu)=(\frac{59}{87}, \frac{28}{87}, \frac{1}{29})$ that can produce special primes in Ford and Maynard's work on prime-producing sieves. Our minimum amount of Type-II information required ($\nu = \frac{1}{29}$) is less than any previous work on this topic using only traditional Type-I and Type-II information.

We investigate the global stability of large solutions to the compressible isentropic Navier-Stokes equations in a three-dimensional (3D) bounded domain with Navier-slip boundary conditions. It is shown that the strong solutions converge to an equilibrium state exponentially in the $L^2$-norm provided the density is essentially uniform-in-time bounded from above. Moreover, we obtain that the density converges to its equilibrium state exponentially in the $L^\infty$-norm if additionally the initial density is bounded away from zero. Furthermore, we derive that the vacuum states will not vanish for any time provided vacuum appears (even at a point) initially. This is the first result concerning the global stability for large strong solutions of compressible Navier-Stokes equations with vacuum in 3D general bounded domains.

The celebrated Borel--Tits theorem provides a classification of abstract isomorphisms between (simple) isotropic groups over fields, showing that such isomorphisms arise from field isomorphisms and group-scheme isomorphisms. In this work, we extend the scope of this classification to certain class of group schemes over arbitrary commutative rings. Specifically, we prove that under suitable conditions abstract isomorphisms between the groups of points of isotropic, absolutely simple, adjoint group schemes over rings admit a description analogous to that in the classical setting: namely, they are induced by isomorphisms of ground rings and isomorphisms of the underlying group schemes. This result generalizes the classical theory to a far broader algebraic context and confirms that the rigidity phenomena observed over fields persist over rings.

Patient-specific computational modeling of the tricuspid valve (TV) is vital for the clinical assessment of heart valve diseases. However, this process is hindered by limitations inherent in the medical image data, such as noise and sparsity, as well as by complex valve dynamics. We present VALVEFIT, a novel GPU-accelerated and differentiable B-spline surface fitting framework that enables rapid reconstruction of smooth, analysis-suitable geometry from point clouds obtained via medical image segmentation. We start with an idealized TV B-spline template surface and optimize its control point positions to fit segmented point clouds via an innovative loss function, balancing shape fidelity and mesh regularization. Novel regularization terms are introduced to ensure that the surface remains smooth, regular, and intersection-free during large deformations. We demonstrate the robustness and validate the accuracy of the framework by first applying it to simulation-derived point clouds that serve as the ground truth. We further show its robustness across different point cloud densities and noise levels. Finally, we demonstrate the performance of the framework toward fitting point clouds obtained from real patients at different stages of valve motion. An isogeometric biomechanical valve simulation is then performed on the fitted surfaces to show their direct applicability toward analysis. VALVEFIT enables automated patient-specific modeling with minimal manual intervention, paving the way for the future development of direct image-to-analysis platforms for clinical applications.

We formulate a geometric version of the Erdős-Hajnal conjecture that applies to finite projective geometries rather than graphs, in both its usual 'induced' form and the multicoloured form. The multicoloured conjecture states, roughly, that a colouring $c$ of the points of $\mathsf{PG}(n-1,q)$ containing no copy of a fixed colouring $c_0$ of $\mathsf{PG}(k-1,q)$ for small $k$ must contain a subspace of dimension polynomial in $n$ that avoids some colour. If $(k,q) = (2,2)$, then $c_0$ is a colouring of a three-element 'triangle', and there are three essentially different cases, all of which we resolve. We derive both the cases where $c_0$ assigns the same colour to two different elements from a recent breakthrough result in additive combinatorics due to Kelley and Meka. We handle the case that $c_0$ is a 'rainbow' colouring by proving that rainbow-triangle-free colourings of projective geometries are exactly those that admit a certain decomposition into two-coloured pieces. This is closely analogous to a theorem of Gallai on rainbow-triangle-free coloured complete graphs. We also show that existing structure theorems resolve certain two-coloured cases where $(k,q) = (2,3)$, and $(k,q) = (3,2)$.

We study the global stability of large solutions to the compressible isentropic magnetohydrodynamic equations in a three-dimensional (3D) bounded domain with Navier-slip boundary conditions. It is shown that the solutions converge to an equilibrium state exponentially in the $L^2$-norm provided the density is essentially uniform-in-time bounded from above. Moreover, we also obtain that the density and magnetic field converge to their equilibrium states exponentially in the $L^\infty$-norm if additionally the initial density is bounded away from zero. These greatly improve the previous work in (J. Differential Equations 288 (2021), 1-39), where the authors considered the torus case and required the $L^6$-norm of the magnetic field to be uniformly bounded as well as zero initial total momentum and an additional restriction $2\mu>\lambda$ for the viscous coefficients. This paper provides the first global stability result for large strong solutions of compressible magnetohydrodynamic equations in 3D general bounded domains.

This letter investigates the integration of pinching-antenna systems (PASS) with orthogonal frequency division multiplexing (OFDM) to ensure their compatibility and to explore the frequency-selective behavior inherent to PASS. First, an end-to-end channel model for OFDM PASS is proposed based on electromagnetic-compliant modeling of waveguides and coupled-mode theory, which includes frequency-dependent waveguide attenuation, dispersion and antenna coupling effect. Furthermore, a critical dependence of the OFDM cyclic prefix (CP) overhead on the proximity of the operating frequency to the waveguide cutoff is revealed. Moreover, the phase misalignment effect across subcarriers in OFDM PASS is derived for an approximate pinching antenna location strategy based on path loss minimization, which reveals the phase misalignment is exacerbated for wider bandwidths and larger array size. Numerical results show that: 1) frequency-selective effects in OFDM PASS lead to substantial variations in subcarrier achievable rates, highlighting the necessity of operating above the waveguide cutoff frequency for effective communications; 2) waveguide dispersion mandates considerable CP overhead when operating near the cutoff frequency, severely impacting the spectral efficiency of OFDM PASS; and 3) the gentle linear waveguide attenuation in a practical PASS significantly more advantageous than the severe logarithmic path loss characteristic of fixed-location antennas.

This paper aims to explore the effect of $\mathbb{Z}_2$-symmetry on grazing-sliding bifurcations in planar Filippov systems. We consider the scenario where the unperturbed system is $\mathbb{Z}_2$-symmetric and its subsystem exhibits a hyperbolic limit cycle grazing the discontinuity boundary at a fold. Employing differential manifold theory, we reveal the intrinsic quantities of unfolding all bifurcations and rigorously demonstrate the emergence of a codimension-two bifurcation under generic $\mathbb{Z}_2$-symmetric perturbations within the Filippov framework. After deriving an explicit non-degenerate condition with respect to parameters, we systematically establish the complete bifurcation diagram with exact asymptotics for all bifurcation boundaries by displacement map method combined with asymptotic analysis.

Let $f^{(r)}(n;s,k)$ be the maximum number of edges in an $n$-vertex $r$-uniform hypergraph containing no $k$ edges on at most $s$ vertices. Brown, Erdős and Sós conjectured in 1973 that the limit $\lim_{n\rightarrow \infty}n^{-2}f^{(3)}(n;k+2,k)$ exists for all $k$. Recently, Delcourt and Postle settled the conjecture and their approach was generalised by Shangguan to every uniformity $r\ge 4$: the limit $\lim_{n\rightarrow \infty}n^{-2}f^{(r)}(n;rk-2k+2,k)$ exists for all $r\ge 3$ and $k\ge 2$. The value of the limit is currently known for $k\in \{2,3,4,5,6,7\}$ due to various results authored by Glock, Joos, Kim, Kühn, Lichev, Pikhurko, Rödl and Sun. In this paper we consider the case $k=8$, determining the value of the limit for each $r\ge 4$ and presenting a lower bound for $k=3$ that we conjecture to be sharp.

Let $\mathcal{F}$ be a singular holomorphic foliation of dimension $k>1$ on a projective $n$-manifold $X$. Assume that the determinant of the normal sheaf of $\mathcal{F}$ is ample (as is always the case when $X=\mathbb{P}^{n}$), and that the singular set $Sing(\mathcal{F})$ has dimension $\leq k-1$. We show that the union of those irreducible components of $Sing(\mathcal{F})$ of dimension exactly $k-1$ is necessarily connected. Consequently, we obtain a Bott-type topological obstruction to the integrability of singular holomorphic distributions, echoing Bott's vanishing theorem, and we answer a question of Cerveau for codimension-one foliations on $\mathbb{P}^{3}$.

We consider a non-exchangeable system of interacting quantum particles with mean-field type interactions, subject to continuous measurement on dense graphs. In the mean-field limit, we derive a graphon-based quantum filtering system, establish its well-posedness, and prove propagation of chaos for multi-class bosonic systems with blockwise interactions. We then discuss applications to quantum state preparation and quantum graphon games.

The decarbonization of energy systems at energy-intensive sites is an essential component of global climate mitigation, yet such transitions involve substantial capital requirements, ongoing technological progress, and the operational complexities of renewable integration. This study presents a dynamic strategic planning framework that applies multi-stage stochastic programming to guide clean electricity transitions at the campus level. The model jointly addresses technology investment, storage operation, and grid interaction decisions while explicitly incorporating uncertainties in future technology cost trajectories and efficiency improvements. By enabling adaptive, stage-wise decision-making, the framework provides a structured approach for large electricity consumers seeking to achieve self-sufficient and sustainable energy systems. The approach is demonstrated through a case study of Middle East Technical University (Ankara, Turkey), which has committed to achieving carbon-neutral electricity by 2040. Through the integration of solar photovoltaics, wind power, and lithium-ion batteries, the model links long-term investment planning with operational-level dynamics by incorporating high-resolution demand and meteorological data. Our findings from the case study, sensitivity analyses, and comparisons with simplified models indicate that accounting for uncertainty and temporal detail is crucial for both the economic viability and operational feasibility of campus-scale clean electricity transitions.

Let $(R,{\frak{m}}_R)$ be a commutative noetherian local ring. Assuming that ${\frak{m}}_R=$$I\oplus J$ is a direct sum decomposition, where $I$ and $J$ are non-zero ideals of $R$, we describe the structure of the Tor algebra of $R$ in terms of the Tor algebras of the rings $R/I$ and $R/J$.

We introduce \emph{hierarchical depth}, a new invariant of line bundles and divisors, defined via maximal chains of effective sub-line bundles. This notion gives rise to \emph{hierarchical filtrations}, refining the structure of the Picard group and providing new insights into the geometry of algebraic surfaces. We establish fundamental properties of hierarchical depth, derive inequalities through intersection theory and the Hodge index theorem, and characterize filtrations that are Hodge-tight. Using this framework, we develop a theory of \emph{hierarchical algebraic geometry codes}, constructed from evaluation spaces along these filtrations. This approach produces nested families of codes with controlled growth of parameters and identifies an optimal intermediate code maximizing a utility function balancing rate and minimum distance. Hierarchical depth thus provides a systematic method to construct AG codes with favorable asymptotic behavior, linking geometric and coding-theoretic perspectives. Our results establish new connections between line bundle theory, surface geometry, and coding theory, and suggest applications to generalized Goppa codes and higher-dimensional evaluation codes.

This paper studies the N-dimensional FENE dumbbell model without velocity dissipation, focusing on the stability and decay of perturbations near the steady solution $(0,\pin)$. Due to the lack of velocity dissipation, the above problems are highly challenging. In fact, without coupling, the corresponding N-dimensional Euler equation near u=0 is well known to be unstable. To overcome this difficulty, we analyze the wave structure arising in the system governing perturbations around the steady state, which originates from the equilibrium configuration and the coupling effects. This wave structure enables us to establish the global stability in the $H^s$-type Sobolev norms. Also, we highlight the critical role of wave structure in the decay estimates of the Euler-FENE dumbbell model. By combining this property with the Fourier splitting method, we derive the decay rate, which is identical to that of the general FENE dumbbell with velocity dissipation.

In this paper, we propose a novel extrapolation coefficient scheme within a new extrapolation term and develop an accelerated proximal gradient algorithm. We establish that the algorithm achieves a sublinear convergence rate. The proposed scheme only requires the Lipschitz constant estimate sequence to satisfy mild initial conditions, under which a key equality property can be derived to support the convergence analysis. Numerical experiments are provided to demonstrate the effectiveness and practical performance of the proposed method.

We study the properties of LP-Sasakian manifolds endowed with generalized $% \eta -$Ricci solitons associated to the general connection. Finally, the existence of such solitons on a 4-dimensional LP-Sasakian manifold is proved by constructing a non-trivial example.

This article deals with the problems of local and global solvability for a semilinear heat equation on the Heisenberg group involving a mixed local and nonlocal nonlinearity. The characteristic features of such equations, arising from the interplay between the geometric structure of the Heisenberg group and the combined nonlinearity, are analyzed in detail. The need to distinguish between subcritical and supercritical regimes is identified and justified through rigorous analysis. On the basis of the study, the author suggests precise conditions under which local-in-time mild solutions exist uniquely for regular, nonnegative initial data. It is proved that global existence holds under appropriate growth restrictions on the nonlinear terms. To complement these results, it is shown, by employing the capacity method, that solutions cannot exist globally in time when the nonlinearity exceeds a critical threshold. As a result, the Fujita exponent is formulated and identified as the dividing line between global existence and finite-time blow-up. In addition, lifespan estimates were obtained in the supercritical regime, providing insight into how the size of the initial data influences the time of blow-up.

A revised proof of the author's earlier result is given. It is shown that a boundary surface-link in the 4-sphere is a ribbon surface-link if the surface-link obtained from it by surgery along a pairwise nontrivial fusion 1-handle system is a ribbon surface-link. As a corollary, the surface-knot obtained from the anti-parallel surface-link of a non-ribbon surface-knot by surgery along a nontrivial fusion 1-handle is a non-ribbon surface-knot. This result answers Cochran's conjecture on non-ribbon sphere-knots in the affirmative.

Simulating electromagnetic fields across broad frequency ranges is challenging due to numerical instabilities at low frequencies. This work extends a stabilized two-step formulation of Maxwell's equations to the time-domain. Using a Galerkin discretization in space, we apply two different time-discretization schemes that are tailored to the first- and second-order in time partial differential equations of the two-step solution procedure used here. To address the low-frequency instability, we incorporate a generalized tree-cotree gauge that removes the singularity of the curl-curl operator, ensuring robustness even in the static limit. Numerical results on academic and application-oriented 3D problems confirm stability, accuracy, and the method's applicability to nonlinear, temperature-dependent materials.

Consider a Chevalley group over a finite field $F_q$ such that the longest element of the Weyl group is central. We construct an involution $\xi\mapsto\xi^!$ of the set of unipotent representations of this group such that the degree polynomial of a unipotent representation $\xi$ is obtained up to sign from the degree polynomial of $\xi^!$ by changing $q$ to $-q$.

In this article we study a class of central extensions of $\mathbb{Z}\wr\mathbb{Z}$, as first described by Hall. On the one hand, we consider groups of this type with cyclic centre, our construction yields a rich class of groups. In particular we obtain a group that is conjugacy separable with solvable word problem but unsolvable conjugacy problem, we obtain a group with large conjugacy separability growth but small conjugator length function and residual finiteness growth, and we also obtain a class of groups that for most functions $f:\mathbb{N}\rightarrow\mathbb{N}$ larger then $n^3$, contain a group $G$ such that the conjugator length of $G$ is given by $f$. On the other hand we also consider a different group with larger centre. This is the first example of a group where the residual finiteness growth is faster than any polynomial but slower than any exponential.

The theory of polynomial-like maps is of fundamental importance in holomorphic dynamics. We study dynamical properties of a larger class of maps. Our main result is that, under some natural conditions, a map of this class has a completely invariant compact set if and only if this set is the filled Julia set of a polynomial-like restriction of the map. We also generalize this result to include maps with non-connected domains of definition.

In this paper, we propose a novel framework, Dynamic Domain Decomposition Physics-Informed Neural Networks (D3PINNs), for solving time-dependent partial differential equations (PDEs). In this framework, solutions of time-dependent PDEs are dynamically captured. First, an approximate solution is obtained by the Physics-Informed Neural Networks (PINNs) containing the domain decomposition, then the time derivative terms in the PDE will be retained and the other terms associated with the solution will be replaced with the approximate solution. As a result, the PDE reduces to an ordinary differential equations (ODEs). Finally, the time-varying solution will be solved by the classical numerical methods for ODEs. D3PINNs retain the computational efffciency and ffexibility inherent to PINNs and enhance the ability for capturing solutions of time-dependent PDEs. Numerical experiments validate the effectiveness of the proposed methods.

Katz and Sarnak conjectured that the behavior of zeros near the central point of any family of $L$-functions is well-modeled by the behavior of eigenvalues near $1$ of some classical compact group (either the symplectic, unitary, or even, odd, or full orthogonal group). In 2018, Knightly and Reno proved that the symmetry group can vary depending on how the $L$-functions in the family are weighted. They observed both orthogonal and symplectic symmetry in the one-level densities of families of cuspidal newform $L$-functions for different choices of weights. We observe the same dependence of symmetry on weights in the $n^{\text{th}}$ centered moments of these one-level densities, for smooth test functions whose Fourier transforms are supported in $\left(-\frac{1}{2n}, \frac{1}{2n}\right)$. To treat the new terms that emerge in our $n$-level calculations when $n>1$, i.e., the cross terms that emerge from $n$-fold products of primes rather than individual primes, we generalize Knightly and Reno's weighted trace formula from primes to arbitrary positive integers. We then perform a delicate analysis of these cross terms to distinguish their contributions to the main and error terms of the $n^{\text{th}}$ centered moments. The final novelty here is an elementary combinatorial trick that we use to rewrite the main number theoretic terms arising from our analysis, facilitating comparisons with random matrix theory.

Let $\mathcal{A}$ be a family of subsets of a finite set. A subfamily of $\mathcal{A}$ is said to be intersecting when any two of its members contain at least one common element. We say that $\mathcal{A}$ is an Erd{\H o}s-Ko-Rado (EKR) family if, for every element $x$ of the set, the subfamily consisting of all members of $\mathcal{A}$ that contain $x$ has the maximum cardinality among all intersecting subfamilies of $\mathcal{A}$. If these subfamilies are the only maximum intersecting subfamilies of $\mathcal{A}$, then $\mathcal{A}$ is called a strong EKR family. In this article, we introduce a compositional framework to establish the EKR and strong EKR properties in set systems when some subfamilies are known to satisfy the EKR or strong EKR properties. Our method is powerful enough to yield simpler proofs for several existing results, including those derived from Katona's cycle method (1968), Borg and Meagher's admissible ordering method (2016), related results on the family of permutations studied by Frankl and Deza (1977) and the family of perfect matchings of complete graphs of even order investigated by Meagher and Moura (2005). To demonstrate the applicability and effectiveness of our method when other existing methods have not been successful, we show that for every fixed $r$-uniform hypergraph $H$ and all sufficiently large integers $n$, the family of all subhypergraphs of the complete $r$-uniform hypergraph on $n$ vertices that are isomorphic to $H$ satisfies the strong EKR property, where two copies of $H$ are considered intersecting if they share at least one common hyperedge. Moreover, when the structural constraint $H$ is restricted to be a cycle, we establish a series of EKR results for families of cycles in the complete graph $K_n$ and the complete bipartite graph $K_{n,n}$ for a broad range of the parameter $n$.

Let $N>1$ and let $\Phi_N(X,Y)\in\mathbb{Z}[X,Y]$ be the modular polynomial which vanishes precisely at pairs of $j$-invariants of elliptic curves linked by a cyclic isogeny of degree $N$. In this note we study the divisibility of the coefficients of $\Phi_N(X+J, Y+J)$ for certain algebraic numbers $J$, in particular $J=0$ and other singular moduli. It turns out that these coefficients are highly divisible by small primes at which $J$ is supersingular.

We show that the Generalized Sato-Tate Conjecture permits to obtain rather precise information on the distribution of the consecutive quadratic residues modulo large primes.

The focus of this work is on the homogeneous and non-homogeneous Dirichlet problem for the Laplacian in bounded Lipschitz domains (BLD). Although it has been extensively studied by many authors, we would like to return to a number of fundamental questions and known results, such as the traces and the maximal regularity of solutions. First, to treat non-homogeneous boundary conditions, we rigorously define the notion of traces for non regular functions. This approach replaces the non-tangential trace notion that has dominated the literature since the 1980s. We identify a functional space E = \{v\in H^{1/2}(\Omega);\nabla v\in [H^1/2(\Omega)]'\} for which the trace operator is continuous from $E$ into $L^2(\Gamma)$. Second, we address the regularity of solutions to the Laplace equation with homogeneous Dirichlet conditions. Using specific equivalent norms in fractional Sobolev spaces and Grisvard's results for polygons and polyhedral domains, we prove that maximal regularity $H^{3/2}$ holds in any BLD $\Omega$, for all right-hand sides in the dual of $H^{1/2}_{00}(\Omega)$. This conclusion contradicts the prevailing claims in the literature since the 1990s. Third, we describe some criteria which establish new uniqueness results for harmonic functions in Lipschitz domains. In particular, we show that if $u\in H^{1/2}(\Omega)$ or $u\in W^{1, 2N/(N+1)}(\Omega)$, is harmonic in $\Omega$ and vanishes on $\Gamma$, then $u= 0$. These criteria play a central role in deriving regularity properties. Finally, we revisit the classical Area Integral Estimate. Using Grisvard's work and an explicit function given by Necas, we show that this inequality cannot hold in its stated form. Since this estimate has been widely used to argue that $H^{3/2}$-regularity is unattainable for data in the dual of $H^{1/2}_{00}(\Omega)$, our counterexample provides a decisive clarification.

We introduce and construct a novel type of canonical metric: the semi-flat constant scalar curvature Kähler (semi-flat cscK) current, which naturally arises in Calabi-Yau fibrations. For a given elliptic surface $X$ with a holomorphic section, We explicitly construct the desired semi-flat cscK current and analyze its behavior along singular parts. We establish its uniqueness under the condition that $X$ possesses at least one singular fiber other than of type $I_b$ or $I_b^*$. These results contribute to a geometric uniformization program for elliptic surfaces.

Nonlinear optimization-based control policies, such as those those arising in nonlinear Model Predictive Control, have seen remarkable success in recent years. These policies require solving computationally demanding nonlinear optimization programs online at each time-step. The resulting solution map, viewed as a function of the measured state of the system and design parameters, may not be differentiable, which poses significant challenges if the control policy is embedded in a gradient-based policy optimization scheme. We propose a principled way to regularize the nonlinear optimization problem, obtaining a surrogate derivative even if when the original problem is not differentiable. The surrogate problem is differentiable by design and its solution map coincides with the solution of the unregularized problem. We demonstrate the effectiveness of our approach in a free-final-time optimal control problem and a receding-horizon nonlinear MPC example.

We prove that the lonely runner conjecture holds for eight runners. Our proof relies on a computer verification and on recent results that allow bounding the size of a minimal counterexample. We note that our approach also applies to the known cases with 4, 5, 6, and 7 runners. We expect that minor improvements to our approach could be enough to solve the cases of 9 or 10 runners.

In this paper, we discuss all the possible pairs $(u,c)\in C(M,\mathbb R)\times\mathbb R$ solving (in the sense of viscosity) the contact Hamilton-Jacobi equation \[ H (x, d_xu, u) = c,\quad x\in M \] of which $M$ is a closed manifold and the continuous Hamiltonian $H: (x,p,u)\in T^*M\times\mathbb R\rightarrow\mathbb R$ is convex, coercive in $p$ but merely non-decreasing in $u$. Firstly, we propose a comparison principle for solutions by using the dynamical information of Mather measures. We then describe the structure of $\mathfrak C$ containing all the $c\in\mathbb R$ makes previous equation solvable. We also propose examples to verify the optimality of our approach.

In-Context Learning (ICL) has emerged as an important new paradigm in natural language processing and large language model (LLM) applications. However, the theoretical understanding of the ICL mechanism remains limited. This paper aims to investigate this issue by studying a particular ICL approach, called concept-based ICL (CB-ICL). In particular, we propose theoretical analyses on applying CB-ICL to ICL tasks, which explains why and when the CB-ICL performs well for predicting query labels in prompts with only a few demonstrations. In addition, the proposed theory quantifies the knowledge that can be leveraged by the LLMs to the prompt tasks, and leads to a similarity measure between the prompt demonstrations and the query input, which provides important insights and guidance for model pre-training and prompt engineering in ICL. Moreover, the impact of the prompt demonstration size and the dimension of the LLM embeddings in ICL are also explored based on the proposed theory. Finally, several real-data experiments are conducted to validate the practical usefulness of CB-ICL and the corresponding theory.

This paper resolves a question of Huneke and Watanabe by proving a sharp upper bound for the multiplicity of Du Bois singularities: at a point of a $d$-dimensional variety with Du Bois singularities and embedding dimension $e$, the multiplicity is at most $\binom{e}{d}$. Additionally, the result recovers the previously known upper bound for the multiplicity of rational singularities.

In this paper, we prove the diameter comparison, the global weighted volume comparison and the splitting theorem in weighted manifolds when the infinity-Bakry-Emery Ricci curvature has a lower bound in the spectrum sense. Our results extend Antonelli-Xu's spectral Bonnet-Myers and Bishop-Gromov theorems, and Antonelli-Pozzetta-Xu's spectral splitting theorem to weighted manifolds. Our results are also some supplements of Chu-Hao's spectral diameter and global volume comparisons, and Yeung's spectral splitting theorem in weighted manifolds.

September 7, 2025 marked the 80th anniversary of the birth of Oleg Marichev. Marichev is known mathematician which has developed many of Mathematica's algorithms for the calculation of definite and indefinite integrals and hypergeometric functions including Meijer G-function.

We consider the following nonlocal critical problem with mixed Dirichlet-Neumann boundary conditions, \begin{equation} \left\{ \begin{array}{ll} (-\Delta)^su=\lambda u+|u|^{2_s^*-2}u &\text{in}\ \Omega,\\ \mkern+38.5mu u=0& \text{on}\ \Sigma_{\mathcal{D}},\\ \mkern+24mu \displaystyle \frac{\partial u}{\partial \nu}=0 &\text{on}\ \Sigma_{\mathcal{N}}, \end{array} \right. \end{equation} where $(-\Delta)^s$, $s\in (1/2,1)$, is the spectral fractional Laplacian operator, $\Omega\subset\mathbb{R}^N$, $N>2s$, is a smooth bounded domain, $2_s^*=\frac{2N}{N-2s}$ denotes the critical fractional Sobolev exponent, $\lambda>0$ is a real parameter, $\nu$ is the outwards normal to $\partial\Omega$, $\Sigma_{\mathcal{D}}$, $\Sigma_{\mathcal{N}}$ are smooth $(N-1)$--dimensional submanifolds of $\partial\Omega$ such that $\Sigma_{\mathcal{D}}\cup\Sigma_{\mathcal{N}}=\partial\Omega$, $\Sigma_{\mathcal{D}}\cap\Sigma_{\mathcal{N}}=\emptyset$ and $\Sigma_{\mathcal{D}}\cap\overline{\Sigma}_{\mathcal{N}}=\Gamma$ is a smooth $(N-2)$--dimensional submanifold of $\partial\Omega$. By employing a $\nabla$-theorem we prove the existence of multiple solutions when the parameter $\lambda$ is in a left neighborhood of a given eigenvalue of $(-\Delta)^s$.

The space of unit flows on a finite acyclic directed graph is a lattice polytope called the flow polytope of the graph. Given a bipartite graph $G$ with minimum degree at least two, we construct two associated acyclic directed graphs: the extension of $G$ and the almost-degree-whiskered graph of $G$. We prove that the normalized volume of the flow polytope for the extension of $G$ is equal to the number of matchings in the almost-degree-whiskered graph of $G$. Further, we refine this result by proving that the Ehrhart $h^*$-polynomial of the flow polytope for the extension of $G$ is equal to the unsigned matching polynomial of the almost-degree-whiskered graph of $G$.

We prove that Hilbert's Tenth Problem over $\mathbb{N}$ remains undecidable when restricted to cubic equations (degree $\leq 3$), resolving the open case $\delta = 3$ identified by Jones (1982) and establishing sharpness against the decidability barrier at $\delta = 2$ (Lagrange's four-square theorem). For any consistent, recursively axiomatizable theory $T$ with Gödel sentence $G_T$, we effectively construct a single polynomial $P(x_1, \ldots, x_m) \in \mathbb{Z}[\mathbf{x}]$ of degree $\leq 3$ such that $T \vdash G_T$ if and only if $\exists \mathbf{x} \in \mathbb{N}^m : P(\mathbf{x}) = 0$. Our reduction proceeds through four stages with explicit degree and variable accounting. First, proof-sequence encoding via Diophantine $\beta$-function and Zeckendorf representation yields $O(KN)$ quadratic constraints, where $K = O(\log(\max_i f_i))$ and $N$ is the proof length. Second, axiom--modus ponens verification is implemented via guard-gadgets wrapping each base constraint $E(\mathbf{x}) = 0$ into the system $u \cdot E(\mathbf{x}) = 0$, $u - 1 - v^2 = 0$, maintaining degree $\leq 3$ while introducing $O(KN^3)$ variables and equations. Third, system aggregation via sum-of-squares merger $P_{\text{merged}} = \sum_{i} P_i^2$ produces a single polynomial of degree $\leq 6$ with $O(KN^3)$ monomials. Fourth, recursive monomial shielding factors each monomial of degree exceeding $3$ in $O(\log d)$ rounds via auxiliary variables and degree-$\leq 3$ equations, adding $O(K^3 N^3)$ variables and restoring degree $\leq 3$. We provide bookkeeping for every guard-gadget and merging operation, plus a unified stage-by-stage variable-count table. Our construction is effective and non-uniform in the uncomputable proof length $N$, avoiding any universal cubic equation. This completes the proof that the class of cubic Diophantine equations over $\mathbb{N}$ is undecidable.

In this paper, we evaluate the following families of definite integrals in closed form and we show that they are expressible only in terms of the dilogarithm function and the inverse tangent integral, and elementary functions. \begin{equation*} \int_{0}^{1}\frac{\log\big(x^m+1\big)}{x+1}\thinspace{\rm d}x \quad \mbox{and}\quad \int_{0}^{1}\frac{\log\big(x^m+1\big)}{x^2+1}\thinspace{\rm d}x, \end{equation*} where $m$ is a positive odd integer. When $m$ is a positive even integer, these integrals have been evaluated previously by Sofo and Batır, and the case where $m$ is an odd integer has been left as open problems. The integrals of the first kind arise in Zagier's work on the Kronecker limit formula. In addition, we demonstrate that a functional equation satisfied by the Herglotz-Zagier-Novikov function is a very specific case of of a more general formula, and give numerous illustrative examples.

We introduce the abstract notion of squarefree-power-like functions, which unify the sequences of squarefree ordinary and symbolic powers of squarefree monomial ideals. By employing the Tor-vanishing criteria for mixed sums of ideals, we establish sharp lower bounds for their Castelnuovo-Mumford regularity in terms of what we call the admissible set of the associated hypergraph. As an application, we derive the first general combinatorial lower bound for the regularity of squarefree symbolic powers of monomial ideals. In the setting of edge ideals, by exploiting the special combinatorial structures of block graphs and Cohen-Macaulay chordal graphs, we show that this bound turns into an exact formula for all squarefree symbolic powers of block graphs, as well as for the second squarefree symbolic powers of edge ideals of Cohen-Macaulay chordal graphs.

We establish fractional Leibniz rules in weighted settings for nonnegative self-adjoint operators on spaces of homogeneous type. Using a unified method that avoids Fourier transforms, we prove bilinear estimates for spectral multiplier on weighted Hardy, Besov and Triebel-Lizorkin spaces. Our approach is flexible and applies beyond the Euclidean setting-covering, for instance, nilpotent Lie groups, Grushin operators, and Hermite expansions-thus extending classical Kato-Ponce inequalities. The framework also yields new weighted bilinear estimates including fractional Leibniz rules for Hermite, Laguerre, and Bessel operator, with applications to scattering formulas and related PDE models.

Recently, Kleshchev and Livesey proved the existence of RoCK $p$-blocks for double covers of symmetric and alternating groups over large enough coefficient rings. They proved that RoCK blocks of double covers are Morita equivalent to standard ``local" blocks via bimodules with endopermutation source. Based on this, Kleshchev and Livesey proved that these RoCK blocks are splendidly Rickard equivalent to their Brauer correspondents. The analogous result for blocks of symmetric groups, a theorem of Chuang and Kessar, was an important step in Chuang and Rouquier ultimately proving Broué's abelian defect group conjecture for symmetric groups. In this paper we show that in most cases the Morita and splendid Rickard equivalences constructed by Kleshchev and Livesey descend to the ring $\mathbb{Z}_p$ of $p$-adic integers, hence prove Kessar and Linckelmann's refinement of Broué's abelian defect group conjecture for most of these RoCK blocks.

How should one define metric space notions of convergence for sequences of spacetimes? Since a Lorentzian manifold does not define a metric space directly, the uniform convergence, Gromov-Hausdorff (GH) convergence, and Sormani-Wenger Intrinsic Flat (SWIF) convergence does not extend automatically. One approach is to define a metric space structure, which is compatible with the Lorentzian structure, so that the usual notions of convergence apply. This approach was taken by C. Sormani and C. Vega when defining the null distance. In this paper, we study sequences of static spacetimes equipped with the null distance under uniform, GH, and SWIF convergence, as well as Hölder bounds. We use the results of the Volume Above Distance Below (VADB) theorem of the author, R. Perales, and C. Sormani to prove an analog of the VADB theorem for sequences of static spacetimes with the null distance. We also give a conjecture of what the VADB theorem should be in the case of sequences of globally hyperbolic spacetimes with the null distance.

Given a $(0,p)$-mixed characteristic complete discrete valued field $\mathcal{K}$ we define a class of finite field extensions called \emph{pseudo-perfect} extensions such that the natural restriction map on the mod-$p$ Milnor $K$-groups is trivial for all $p\neq 2$. This implies that pseudo-perfect extensions split every element in $H^i(\mathcal{K},\mu_p^{\otimes i-1})$ yielding period-index bounds for Brauer classes as well as higher cohomology classes of $\mathcal{K}$. As a corollary, we prove a conjecture of Bhaskhar-Haase that the Brauer $p$-dimension of $\mathcal{K}$ is upper bounded by $n+1$ where $n$ is the $p$-rank of the residue field. When $\mathcal{K}$ is the fraction field of a complete regular ring, we show that any $p$-torsion element in $Br(\mathcal{K})$ that is nicely ramified is split by a pseudo-perfect extension yielding a bound on its index. We then use patching techniques of Harbater, Hartmann and Krashen to show that the Brauer $p$-dimension of semi-global fields of residual characteristic $p$ is at most $n+2$ and also give uniform $p$-bounds for higher cohomologies. These bounds are sharper than previously known in the work of Parimala-Suresh

Consider a Chevalley group over a finite field F_q such that the longest element in the Weyl group is central. In this paper we study the effect of changing q to -q in the polynomials which give the character values of unipotent representations of our group at semisimple elements.

We study Dirichlet series arising as linear functionals on an inner product space of meromorphic functions and establish a relation between the discontinuities of the former on the boundary and the poles and zeros of the latter on the imaginary axis. As an example application of Delange's Tauberian theorem, it is shown that the conjectured asymptotic in the additive divisor problem follows conditionally on the non-vanishing of a certain meromorphic function and its first derivative on the imaginary axis.

This article addresses the inverse source problem for a nonlocal heat equation involving the fractional Laplacian. The primary goal is to reconstruct the spatial component of the source term from partial observations of the system's state and its time derivative over a subset of the domain. A reconstruction formula for the Fourier coefficients of the unknown source is derived, leveraging the null controllability property of the fractional heat equation when the fractional order lies in the interval $s\in(1/2,1)$. The methodology builds on spectral analysis and Volterra integral equations, providing a robust framework for recovering spatial sources under limited measurement data. Numerical experiments confirm the accuracy and stability of the proposed approach.

Brownian Boost is a one-parameter family of stochastic differential games played on the real line in which players spend at rates of their choosing in an ongoing effort to influence the drift of a randomly diffusing point particle~$X$. One or other player is rewarded, at time infinity, according to whether~$X$ tends to plus or minus infinity. Each player's net receipt is the final reward (only for the victor) minus the player's total spend. We characterise and explicitly compute the time-homogeneous Markov-perfect Nash equilibria of Brownian Boost, finding the derivatives of the players' expected payoffs to solve a pair of coupled first-order non-linear ODE. Brownian Boost is a high-noise limit of a two-dimensional family of player-funded tug-of-war games, one of which was studied in~\cite{LostPennies}. We analyse the discrete games, finding them, and Brownian Boost, to exemplify key features studied in the economics literature of tug-of-war initiated by~\cite{HarrisVickers87}: a battlefield region where players spend heavily; stakes that decay rapidly but asymmetrically in distance to the battlefield; and an effect of discouragement that makes equilibria fragile under asymmetric perturbation of incentive. Tug-of-war has a parallel mathematical literature derived from~\cite{PSSW09}, which solved the scaled fair-coin game in a Euclidean domain via the infinity Laplacian PDE. By offering an analytic solution to Brownian Boost, a game that models strategic interaction and resource allocation, we seek to build a bridge between the two tug-of-war literatures.

Let $p$ be a prime $\equiv 3$ mod 4, $p>3$, and suppose that 10 has the order $(p-1)/2$ mod p. Then $1/p$ has a decimal period of length $(p-1)/2$. We express the frequency of each digit $0,\ldots,9$ in this period in terms of the class numbers of two imaginary quadratic number fields. We also exhibit certain analogues of this result, so for the case that 10 is a primitive root mod $p$ and for octal digits of $1/p$.

We consider the two-cardinal Kurepa Hypothesis $\mathsf{KH}(\kappa,\lambda)$. We observe that if $\kappa\leq\lambda<\mu$ are infinite cardinals then $\lnot\mathsf{KH}(\kappa,\lambda)\land\mathsf{KH}(\kappa,\mu)\rightarrow\mathsf{KH}(\lambda^+,\mu)$, and show that in some sense this is the only $\mathsf{ZFC}$ constraint. The case of singular $\lambda$ and its relation to Chang's Conjecture and scales is discussed. We also extend an independence result about Kurepa and Aronszajn trees due to Cummings to the case of successors of singular cardinal.

A character table $X$ for a saturated fusion system $\mathcal{F}$ on a finite $p$-group $S$ is the square matrix of values associated to a basis of virtual $\mathcal{F}$-stable ordinary characters of $S$. We investigate a conjecture of the second author which equates the $p$-part of $|$det$(X)|^2$ with the product of the orders of $S$-centralisers of fully $\mathcal{F}$-centralised $\mathcal{F}$-class representatives. This statement is exactly column orthogonality for the character table of $S$ when $\mathcal{F}=\mathcal{F}_S(S)$. We prove the conjecture when $\mathcal{F}=\mathcal{F}_S(G)$ is realised by some finite group $G$ with Sylow $p$-subgroup $S$, and for all simple fusion systems when $|S| \le p^4$.

We show that the natural map from the syntomification of a ring $R$ to the stack of $R$-algebra stacks is fully faithful, answering a question of Drinfeld, and we describe its essential image in terms of underlying monoid stacks. We also give similar statements in the characteristic 0 filtered de Rham, $\ell = p$ étale, and Betti settings.

Restoring power distribution systems after extreme events such as tornadoes presents significant logistical and computational challenges. The complexity arises from the need to coordinate multiple repair crews under uncertainty, manage interdependent infrastructure failures, and respect strict sequencing and routing constraints. Existing methods often rely on deterministic heuristics or simplified models that fail to capture the interdependencies between power and transportation networks, do not adequately model uncertainty, and lack representation of the interrelated dynamics and dependencies among different types of repair crews--leading to suboptimal restoration outcomes. To address these limitations, we develop a stochastic two-stage mixed-integer programming framework for proactive crew allocation, assignment, and routing in power grid restoration. The primary objective of our framework is to minimize service downtime and enhance power restoration by efficiently coordinating repair operations under uncertainty. Multiple repair crews are modeled as distinct agents, enabling decentralized coordination and efficient task allocation across the network. To validate our approach, we conduct a case study using the IEEE 8500-node test feeder integrated with a real transportation network from the Dallas-Fort Worth (DFW) region. Additionally, we use tornado event data from the DFW area to construct realistic failure scenarios involving damaged grid components and transportation links. Results from our case study demonstrate that the proposed method enables more coordinated and efficient restoration strategies. The model facilitates real-time disaster response by supporting timely and practical power grid restoration, with a strong emphasis on interoperability and crew schedule coordination.

We revisit the results of Kitson and Timoney \emph{[J.~Math.~Anal.~Appl.~\textbf{378} (2011), 680--686]} on the spaceability of complements of operator ranges, extending one of their main theorems to the general Fréchet setting. In particular, we provide an affirmative answer to the question posed in \emph{Remark~3.4} of that paper, showing that the conclusion remains valid when the operators act between Fréchet spaces. Moreover, we show that the same phenomenon occurs for arbitrary (possibly uncountable) families of operators. The arguments presented here follow the spirit of the original work.

The Sinc approximation is known to be a highly efficient approximation formula for rapidly decreasing functions. For unilateral rapidly decreasing functions, which rapidly decrease as $x\to\infty$ but does not as $x\to-\infty$, an appropriate variable transformation makes the functions rapidly decreasing. As such a variable transformation, Stenger proposed $t = \sinh(\log(\operatorname{arsinh}(\exp x)))$, which enables the Sinc approximation to achieve root-exponential convergence. Recently, another variable transformation $t = 2\sinh(\log(\log(1+\exp x)))$ was proposed, which improved the convergence rate. Furthermore, its computational error bound was provided. However, this improvement was not significant because the convergence rate remained root-exponential. To improve the convergence rate significantly, this study proposes a new transformation, $t = 2\sinh(\log(\log(1+\exp(\pi\sinh x))))$, which is categorized as the double-exponential (DE) transformation. Furthermore, this study provides its computational error bound, which shows that the proposed approximation formula can achieve almost exponential convergence. Numerical experiments that confirm the theoretical result are also provided.

We establish the preservation of the way-below relation with respect to the tensor product.

A paper torus is a piecewise linear isometric embedding of a flat torus into $\R^3$. Following up on the $8$-vertex paper tori discovered by the second author, we prove universality and collapsibility results about these objects. One corollary is that any flat torus without reflection symmetry is realized as an $8$-vertex paper torus. Another corollary is that, for any $\epsilon>0$, there is an $8$-vertex paper torus within $\epsilon$ of a unit equilateral triangle in the Hausdorff metric.

In prior work, we showed that subsets of $\mathbb{F}_{p}^{n}$ of $\mathrm{VC_{2}}$-dimension at most $k$ are well approximated by a union of atoms of a quadratic factor of complexity $(\ell,q)$, where the complexity $\ell$ of the linear part and the complexity $q$ of the quadratic part are both bounded in terms of $k$, $p$, and the desired level of approximation $\mu$. A key tool in the proof of this result was an arithmetic regularity lemma for the Gowers $U^3$-norm by Green and Tao, which resulted in tower-type bounds (in terms of $\mu^{-1}$) on both $\ell$ and $q$. In the present paper we show that for sets of bounded $\mathrm{VC}_2$-dimension, the bound on $q$ can be substantially improved. Specifically, we will prove that any set $A\subseteq G=\mathbb{F}_p^n$ of $\mathrm{VC}_2$-dimension at most $k$ is approximately equal (up to error $\mu |G|$) to a union of atoms of a quadratic factor whose quadratic complexity is at most $\log_p(\mu^{-k-o(1)})$, implying that the purely quadratic component of the factor partitions the group into $\mu^{-k-o(1)}$ many parts. We achieve this by using our earlier result to obtain an initial quadratic factor $\mathcal{B}$, and then applying a generalization of an argument of Alon, Fox and Zhao for subsets of $\mathbb{F}_{p}^{n}$ of bounded $\mathrm{VC}$-dimension to the label space (also known as "configuration space") of $\mathcal{B}$. A related strategy was employed in earlier work of the authors on $\mathrm{NFOP}_2$ subsets of $\mathbb{F}_p^n$, and in work of the first author in the context of 3-uniform hypergraphs.

We show that a subset of $\mathbb{F}_{p}^{n}$ of $\mathrm{VC_{2}}$-dimension at most $k$ is well approximated by a union of atoms of a quadratic factor of complexity $(\ell,q)$ (denoting the complexities of the linear and quadratic part, respectively), where $\ell$ and $q$ are bounded by a constant depending only on $k$ and the desired level of approximation. This generalises a result of Alon, Fox and Zhao on the structure of sets of bounded $\mathrm{VC}$-dimension, and is analogous to contemporaneous work of the authors arXiv:2111.01737 in the setting of 3-uniform hypergraphs. The main result originally appeared--albeit with a different proof--in a 2021 preprint arXiv:2111.01739, which has since been split into two: the present work, which focuses on higher arity NIP and develops a theory of local uniformity semi-norms of possibly independent interest, and its companion arXiv:2111.01739, which strengthens these results under a generalized notion of stability.

Moduli spaces of instantons on ALE spaces for classical groups are examples of fixed point sets of involutions on quiver varieties, i.e., $\sigma$-quiver varieties. In 2018 Yiqiang Li considered their equivariant cohomology, and by stable envelope of Maulik-Okounkov, constructed representations of coideal subalgebras of Maulik-Okounkov Yangian, called twisted Yangian. We calculate $K$-matrices as matrices in examples, identified the twisted Yangians with ones studied in other literature, and clarify conditions which we should impose to make them well-defined.

We present some review material relating to the topic of optimal asymptotic expansions of correlation functions and associated observables for $\beta$ ensembles in random matrix theory. We also give an introduction to a related line of study that we are presently undertaking.

In the present paper, we generalize the notion of the $p$-Bergman kernel and the $\xi$-Bergman kernel to the $p$-Bergman kernel with respect to a functional $\xi$, and establish some properties of the $p$-Bergman kernel with respect to $\xi$. We also study the relations between the $L^p$ versions of higher order Bergman kernels and $\xi$-Bergman kernels, and as applications we give the reproofs and generalizations of some previous results of Blocki and Zwonek about higher order Bergman kernels.

The aim of this paper is to investigate the spectral theory of unimodular random graphs and graphings representing them. We prove that Bernoulli graphings are relatively Ramanujan with respect to their skeleton Markov chain. That is, the part of their spectrum that comes from the random labels falls within the appropriate Alon-Boppana bound. This result complements an example due to Frączyk of an ergodic unimodular random graph with almost sure spectral gap but non-expanding Bernoulli graphing. We also highlight connections of our work with the theory of finite random graphs. Exploiting the result of Bordenave and Collins on random lifts being relatively almost Ramanujan, we prove a strengthening of our main theorem for unimodular quasi-transitive quasi-trees.

In this paper, we describe the necessary procedures for accurately simulating digital wireless communication systems operating in the mediumband, aimed at both beginners and experts. In the research literature, digital wireless communication systems are typically simulated in the discrete-time complex baseband domain, where pulse shaping, upconversion, mixing, carrier synchronization, and symbol timing synchronization are often ignored. These assumptions are indeed sufficient in most cases, but to capture the essence of communication in the mediumband, certain physical layer (PHY) operations should be simulated in detail. In this paper, we concisely describe how to simulate a mediumband wireless communication scenario from a single transmitter (TX) to a single receiver (RX) in MATLAB, elaborating the operation of key PHY subsystems. The approach described here ensures that the simulated system captures the delicate dynamics of mediumband wireless communication, including the effect of deep fading avoidance.

We construct two counterexamples that resolve long-standing open problems on greedy approximation theory with respect to bases, posed in [F. Albiac et al., Dissertationes Math. 560 (2021)] and restated in [F. Albiac, J. L. Ansorena, V. Temlyakov, J. Approx. Theory 307 (2025)]. Our first result exhibits a quasi-Banach space $\mathbb{X}$ with an almost greedy basis which, when transported to the Banach envelope of $\mathbb{X}$, ceases to be quasi-greedy. This shows that the passage to the Banach envelope, although it preserves linear and lattice structure, may radically disrupt the performance of the thresholding greedy algorithm, to the extent that in some respects it could perform better in a quasi-Banach space than in its Banach envelope. Our second result constructs an almost greedy Markushevich basis in a nonlocally convex quasi-Banach space $\mathbb{Y}$ which fails to be a Schauder basis under any reordering. Together, these examples highlight that local convexity and the Banach envelope construction play an unexpectedly active role in shaping greedy approximation phenomena, revealing structural differences between Banach and quasi-Banach spaces that go beyond the classical theory of bases.

In this paper, we uncover a new uncertainty principle that governs the complexity of Boolean functions. This principle manifests as a fundamental trade-off between two central measures of complexity: a combinatorial complexity of its supported set, captured by its Vapnik-Chervonenkis dimension ($\mathrm{VC}(f)$), and its algebraic structure, captured by its polynomial degree over various fields. We establish two primary inequalities that formalize this trade-off: $\mathrm{VC}(f)+\mathrm{deg}(f)\ge n,$ and $\mathrm{VC}(f)+\mathrm{deg}_{\mathbb{F}_2}(f)\ge n$. In particular, these results recover the classical uncertainty principle on the discrete hypercube, as well as the Sziklai--Weiner's bound in the case of $\mathbb{F}_2$.

A separable quantum state shared between parties $A$ and $B$ can be symmetrically extended to a quantum state shared between party $A$ and parties $B_1,\ldots ,B_k$ for every $k\in\mathbf{N}$. Quantum states that are not separable, i.e., entangled, do not have this property. This phenomenon is known as "monogamy of entanglement". We show that monogamy is not only a feature of quantum theory, but that it characterizes the minimal tensor product of general pairs of convex cones $\mathsf{C}_A$ and $\mathsf{C}_B$: The elements of the minimal tensor product $\mathsf{C}_A\otimes_{\min} \mathsf{C}_B$ are precisely the tensors that can be symmetrically extended to elements in the maximal tensor product $\mathsf{C}_A\otimes_{\max} \mathsf{C}^{\otimes_{\max} k}_B$ for every $k\in\mathbf{N}$. Equivalently, the minimal tensor product of two cones is the intersection of the nested sets of $k$-extendible tensors. It is a natural question when the minimal tensor product $\mathsf{C}_A\otimes_{\min} \mathsf{C}_B$ coincides with the set of $k$-extendible tensors for some finite $k$. We show that this is universally the case for every cone $\mathsf{C}_A$ if and only if $\mathsf{C}_B$ is a polyhedral cone with a base given by a product of simplices. Our proof makes use of a new characterization of products of simplices up to affine equivalence that we believe is of independent interest.

We present a streamlined and simplified exponential lower bound on the length of proofs in intuitionistic implicational logic, adapted to Gordeev and Haeusler's dag-like natural deduction.

Quantum information processing and computing tasks can be understood as quantum networks, comprising quantum states and channels and possible physical transformations on them. It is hence pertinent to estimate the change in informational content of quantum processes due to physical transformations they undergo. The physical transformations of quantum states are described by quantum channels, while the transformations of quantum channels are described by quantum superchannels. In this work, we determine fundamental limitations on how well the physical transformation on quantum channels can be undone or reversed, which are of crucial interest to design and benchmark quantum information and computation devices. In particular, we refine (strengthen) the quantum data processing inequality for quantum channels under the action of quantum superchannels. We identify a class of quantum superchannels, which appears to be the superchannel analogue of subunital quantum channels, under the action of which the entropy of an arbitrary quantum channel is nondecreasing. We also provide a refined inequality for the entropy change of quantum channels under the action of an arbitrary quantum superchannel.

We study an $\ell_{1}$-regularized generalized least-squares (GLS) estimator for high-dimensional regressions with autocorrelated errors. Specifically, we consider the case where errors are assumed to follow an autoregressive process, alongside a feasible variant of GLS that estimates the structure of this process in a data-driven manner. The estimation procedure consists of three steps: performing a LASSO regression, fitting an autoregressive model to the realized residuals, and then running a second-stage LASSO regression on the rotated (whitened) data. We examine the theoretical performance of the method in a sub-Gaussian random-design setting, in particular assessing the impact of the rotation on the design matrix and how this impacts the estimation error of the procedure. We show that our proposed estimators maintain smaller estimation error than an unadjusted LASSO regression when the errors are driven by an autoregressive process. A simulation study verifies the performance of the proposed method, demonstrating that the penalized (feasible) GLS-LASSO estimator performs on par with the LASSO in the case of white noise errors, whilst outperforming when the errors exhibit significant autocorrelation.

Observational studies are valuable tools for inferring causal effects in the absence of controlled experiments. However, these studies may be biased due to the presence of some relevant, unmeasured set of covariates. One approach to mitigate this concern is to identify hypotheses likely to be more resilient to hidden biases by splitting the data into a planning sample for designing the study and an analysis sample for making inferences. We devise a powerful and flexible method for selecting hypotheses in the planning sample when an unknown number of outcomes are affected by the treatment, allowing researchers to gain the benefits of exploratory analysis and still conduct powerful inference under concerns of unmeasured confounding. We investigate the theoretical properties of our method and conduct extensive simulations that demonstrate pronounced benefits, especially at higher levels of allowance for unmeasured confounding. Finally, we demonstrate our method in an observational study of the multi-dimensional impacts of a devastating flood in Bangladesh.

This paper introduces the Kernel Neural Operator (KNO), a provably convergent operator-learning architecture that utilizes compositions of deep kernel-based integral operators for function-space approximation of operators (maps from functions to functions). The KNO decouples the choice of kernel from the numerical integration scheme (quadrature), thereby naturally allowing for operator learning with explicitly-chosen trainable kernels on irregular geometries. On irregular domains, this allows the KNO to utilize domain-specific quadrature rules. To help ameliorate the curse of dimensionality, we also leverage an efficient dimension-wise factorization algorithm on regular domains. More importantly, the ability to explicitly specify kernels also allows the use of highly expressive, non-stationary, neural anisotropic kernels whose parameters are computed by training neural networks. Numerical results demonstrate that on existing benchmarks the training and test accuracy of KNOs is comparable to or higher than popular operator learning techniques while typically using an order of magnitude fewer trainable parameters, with the more expressive kernels proving important to attaining high accuracy. KNOs thus facilitate low-memory, geometrically-flexible, deep operator learning, while retaining the implementation simplicity and transparency of traditional kernel methods from both scientific computing and machine learning.

Let $\Omega\subset \mathbb{R}^d$ be a bounded domain. We consider the problem of how efficiently shallow neural networks with the ReLU$^k$ activation function can approximate functions from Sobolev spaces $W^s(L_p(\Omega))$ with error measured in the $L_q(\Omega)$-norm. Utilizing the Radon transform and recent results from discrepancy theory, we provide a simple proof of nearly optimal approximation rates in a variety of cases, including when $q\leq p$, $p\geq 2$, and $s \leq k + (d+1)/2$. The rates we derive are optimal up to logarithmic factors, and significantly generalize existing results. An interesting consequence is that the adaptivity of shallow ReLU$^k$ neural networks enables them to obtain optimal approximation rates for smoothness up to order $s = k + (d+1)/2$, even though they represent piecewise polynomials of fixed degree $k$.

We present results on the geometry of the positive equilibrium set of a mass action network. Any mass action network gives rise to a parameterised family of polynomial equations whose positive solutions are the positive equilibria of the network. Here, we start by deriving alternative systems of equations, whose solutions are in smooth, one-to-one correspondence with positive equilibria of the network, and capture degeneracy or nondegeneracy of the corresponding equilibria. The derivation leads us to consider partitions of networks in a natural sense, and we explore the implications of choosing different partitions. The alternative systems are often simpler than the original mass action equations, sometimes giving explicit parameterisations of positive equilibria, and allowing us to rapidly identify various algebraic and geometric properties of the positive equilibrium set, including toricity and local toricity. We can use the approaches we develop to bound the number of positive nondegenerate equilibria on stoichiometric classes; to derive semialgebraic descriptions of the parameter regions for multistationarity; and to study bifurcations. We present the main construction, various consequences for particular classes of networks, and numerous examples. We also develop additional techniques specifically for quadratic networks, the most common class of networks in applications, and use these techniques to derive strengthened results for quadratic networks.

A system $\boldsymbol\lambda_{\theta}$ is developed that combines modal logic and simply-typed lambda calculus, and that generalizes the system studied by Montague and Gallin. Whereas Montague and Gallin worked with Church's simple theory of types, the system $\boldsymbol\lambda_{\theta}$ is developed in the typed base theory most commonly used today, namely the simply-typed lambda calculus. Further, the system $\boldsymbol\lambda_{\theta}$ is controlled by a parameter $\theta$ which allows more options for state types and state variables than is present in Montague and Gallin. A main goal of the paper is to establish the basic metatheory of $\boldsymbol\lambda_{\theta}$: (i) a completeness theorem is proven for $\beta\eta$-reduction, and (ii) an Andrews-like characterization of Henkin models in terms of combinatory logic is given; and this involves, with some necessity, a distanced version of $\beta$-reduction and a $\mathsf{BCKW}$-like basis rather than $\mathsf{SKI}$-like basis. Further, conservation of the maximal system $\boldsymbol\lambda_{\omega}$ over $\boldsymbol\lambda_{\theta}$ is proven, and expressibility of $\boldsymbol\lambda_{\omega}$ in $\boldsymbol\lambda_{\theta}$ is proven; thus these modal logics are highly expressive. Similar results are proven for the relation between $\boldsymbol\lambda_{\omega}$ and $\boldsymbol\lambda$, the corresponding ordinary simply-typed lambda calculus. This answers a question of Zimmermann in the simply-typed setting. In a companion paper this is extended to Church's simple theory of types.

Motivated by the problem of multi-twist operators in general CFTs, we study the leading-twist states of the $N$-body problem in AdS at large spin $J$. We find that for the majority of states the effective quantum-mechanical problem becomes semiclassical with $\hbar=1/J$. The classical system at $J=\infty$ has $N-2$ degrees of freedom, and the classical phase space is identified with the positive Grassmannian $\mathrm{Gr}_{+}(2,N)$. The quantum problem is recovered via a Berezin-Toeplitz quantization of a classical Hamiltonian, which we describe explicitly. For $N=3$ the classical system has one degree of freedom and a detailed structure of the spectrum can be obtained from Bohr-Sommerfeld conditions. For all $N$, we show that the lowest excited states are approximated by a harmonic oscillator and find explicit expressions for their energies.

We study the operator product expansion (OPE) of identical scalars in a conformal four-point correlator as a Stieltjes moment problem, and use Riemann-Liouville type fractional differential operators to generate classical moments from the correlation function. We use crossing symmetry to derive leading and subleading relations between moments in $\Delta$ and $J_2 \equiv \ell(\ell+d-2)$ in the ``heavy" limit of large external scaling dimension, and combine them with constraints from unitarity to derive two-sided bounds on moment sequences in $\Delta$ and the covariance between $\Delta$ and $J_2$. The moment sequences which saturate these bounds produce ``saddle point" solutions to the crossing equations which we identify as particular limits of correlators in a generalized free field (GFF) theory. This motivates us to study perturbations of heavy GFF four-point correlators by way of saddle point analysis, and we show that saddles in the OPE arise from contributions of fixed-length operator families encoded by a decomposition into higher-spin conformal blocks. To apply our techniques, we consider holographic correlators of four identical single scalar fields perturbed by a bulk interaction, and use their first few moments to derive Gaussian weight-interpolating functions that predict the OPE coefficients of interacting double-twist operators in the heavy limit.

Breathers have been experimentally and theoretically found in many physical systems -- in particular, in integrable nonlinear-wave models. A relevant problem is to study the \textit{breather gas}, which is the limit, for $N\rightarrow \infty $, of $N$-breather solutions. In this paper, we investigate the breather gas in the framework of the focusing nonlinear Schrödinger (NLS) equation with nonzero boundary conditions, using the inverse scattering transform and Riemann-Hilbert problem. We address aggregate states in the form of $N$-breather solutions, when the respective discrete spectra are concentrated in specific domains. We show that the breather gas coagulates into a single-breather solution whose spectral eigenvalue is located at the center of the circle domain, and a multi-breather solution for the higher-degree quadrature concentration domain. These coagulation phenomena in the breather gas are called \textit{breather shielding}. In particular, when the nonzero boundary conditions vanish, the breather gas reduces to an $n$-soliton solution. When the discrete eigenvalues are concentrated on a line, we derive the corresponding Riemann-Hilbert problem. When the discrete spectrum is uniformly distributed within an ellipse, it is equivalent to the case of the line domain. These results may be useful to design experiments with breathers in physical settings.

We develop a model predictive control (MPC) policy for station keeping on a Near-Rectilinear Halo Orbit (NRHO). The proposed policy achieves full-state tracking of a reference NRHO via a multiple-maneuver control horizon, each spaced one revolution apart to abide by typical mission operation requirements. We prove that the proposed policy is recursively feasible, and perform numerical evaluation in an output-feedback setting by incorporating a navigation filter and realistic operational uncertainties, where the proposed MPC is compared against the state-of-the-art station-keeping algorithm adopted for the Gateway. Our approach successfully maintains the spacecraft in the vicinity of the reference NRHO at a similar cumulative cost as existing station-keeping methods without encountering phase deviation issues, a common drawback of existing methods with one maneuver per revolution.

Inspired by the recent successes of Inverse Optimization (IO) across various application domains, we propose a novel offline Reinforcement Learning (ORL) algorithm for continuous state and action spaces, leveraging the convex loss function called ``sub-optimality loss" from the IO literature. To mitigate the distribution shift commonly observed in ORL problems, we further employ a robust and non-causal Model Predictive Control (MPC) expert steering a nominal model of the dynamics using in-hindsight information stemming from the model mismatch. Unlike the existing literature, our robust MPC expert enjoys an exact and tractable convex reformulation. In the second part of this study, we show that the IO hypothesis class, trained by the proposed convex loss function, enjoys ample expressiveness and achieves competitive performance comparing with the state-of-the-art (SOTA) methods in the low-data regime of the MuJoCo benchmark while utilizing three orders of magnitude fewer parameters, thereby requiring significantly fewer computational resources. To facilitate the reproducibility of our results, we provide an open-source package implementing the proposed algorithms and the experiments.

Exceptional points (EPs) in non-Hermitian photonic systems have attracted considerable research interest due to their singular eigenvalue topology and associated anomalous physical phenomena. These properties enable diverse applications ranging from enhanced quantum metrology to chiral light-matter interactions. Practical implementation of high order EPs in optical platforms however remains fundamentally challenging, requiring precise multi-parameter control that often exceeds conventional design capabilities. This work presents a novel framework for engineering high order EPs through transformation optics (TO) principles, establishing a direct correspondence between mathematical singularities and physically controllable parameters. Our TO-based paradigm addresses critical limitations in conventional Hamiltonian approaches, where abstract parameter spaces lack explicit connections to experimentally accessible degrees of freedom, while simultaneously providing full-field mode solutions. In contrast to prevailing parity-time-symmetric architectures, our methodology eliminates symmetry constraints in EP design, significantly expanding the possibilities in non-Hermitian photonic engineering. The proposed technique enables unprecedented control over EP formation and evolution in nanophotonic systems, offering new pathways for developing topological optical devices with enhanced functionality and robustness.

We investigate the gravitational field of a kinetic gas beyond its usual derivation from the second moment of the one-particle distribution function (1PDF), that serves as energy-momentum tensor in the Einstein equations. This standard procedure raises the question why the other moments of the 1PDF (which are needed to fully characterize the kinematical properties of the gas) do not contribute to the gravitational field and what could be their relevance in addressing the dark energy problem? Using the canonical coupling of the entire 1PDF to Finsler spacetime geometry via the Finsler gravity equation, we show that these higher moments contribute non-trivially. A Finslerian geometric description of our universe allows us to determine not only the scale factor but also of the causal structure dynamically. We find that already a Finslerian vacuum solution naturally permits an exponential expanding universe, without the need for a cosmological constant or any additional quantities. This solution possesses a causal structure which is a mild deformation of the causal structure of Friedmann-Lemaître-Robertson-Walker (FLRW) geometry; close to the rest frame defined by cosmological time (i.e., for slowly moving objects), the causal structures of the two geometries are nearly indistinguishable.

Information processing in the brain is coordinated by the dynamic activity of neurons and neural populations at a range of spatiotemporal scales. These dynamics, captured in the form of electrophysiological recordings and neuroimaging, show evidence of time-irreversibility and broken detailed balance suggesting that the brain operates in a nonequilibrium stationary state. Furthermore, the level of nonequilibrium, measured by entropy production or irreversibility appears to be a crucial signature of cognitive complexity and consciousness. The subsequent study of neural dynamics from the perspective of nonequilibrium statistical physics is an emergent field that challenges the assumptions of symmetry and maximum-entropy that are common in traditional models. In this review, we discuss the plethora of exciting results emerging at the interface of nonequilibrium dynamics and neuroscience. We begin with an introduction to the mathematical paradigms necessary to understand nonequilibrium dynamics in both continuous and discrete state-spaces. Next, we review both model-free and model-based approaches to analysing nonequilibrium dynamics in both continuous-state recordings and neural spike-trains, as well as the results of such analyses. We briefly consider the topic of nonequilibrium computation in neural systems, before concluding with a discussion and outlook on the field.

We prove the invariance of scalar Feynman graphs of any planar topology under the Yangian level-one momentum symmetry given certain constraints on the propagator powers. The proof relies on relating this symmetry to a planarized version of the conformal simplices of Bzowski, McFadden and Skenderis. In particular, this proves a momentum-space analogue of the position-space conformal condition on propagator powers. When combined with the latter, the invariance under the level-one momentum implies full Yangian symmetry of the considered graphs. These include all scalar Feynman integrals for which a Yangian symmetry was previously demonstrated at the level of examples, e.g. the fishnet or loom graphs, as well as generalizations to graphs with massive propagators.

The widespread utilization of language models in modern applications is inconceivable without Parameter-Efficient Fine-Tuning techniques, such as low-rank adaptation ($\texttt{LoRA}$), which adds trainable adapters to selected layers. Although $\texttt{LoRA}$ may obtain accurate solutions, it requires significant memory to train large models and intuition on which layers to add adapters. In this paper, we propose a novel method, $\texttt{WeightLoRA}$, which overcomes this issue by adaptive selection of the most critical $\texttt{LoRA}$ heads throughout the optimization process. As a result, we can significantly reduce the number of trainable parameters while maintaining the capability to obtain consistent or even superior metric values. We conduct experiments for a series of competitive benchmarks and DeBERTa, BART, and Llama models, comparing our method with different adaptive approaches. The experimental results demonstrate the efficacy of $\texttt{WeightLoRA}$ and the superior performance of $\texttt{WeightLoRA+}$ in almost all cases.

Fine-tuning Large Language Models (LLMs) is essential for adapting pre-trained models to downstream tasks. Yet traditional first-order optimizers such as Stochastic Gradient Descent (SGD) and Adam incur prohibitive memory and computational costs that scale poorly with model size. In this paper, we investigate zero-order (ZO) optimization methods as a memory- and compute-efficient alternative, particularly in the context of parameter-efficient fine-tuning techniques like LoRA. We propose $\texttt{JAGUAR SignSGD}$, a ZO momentum-based algorithm that extends ZO SignSGD, requiring the same number of parameters as the standard ZO SGD and only $\mathcal{O}(1)$ function evaluations per iteration. To the best of our knowledge, this is the first study to establish rigorous convergence guarantees for SignSGD in the stochastic ZO case. We further propose $\texttt{JAGUAR Muon}$, a novel ZO extension of the Muon optimizer that leverages the matrix structure of model parameters, and we provide its convergence rate under arbitrary stochastic noise. Through extensive experiments on challenging LLM fine-tuning benchmarks, we demonstrate that the proposed algorithms meet or exceed the convergence quality of standard first-order methods, achieving significant memory reduction. Our theoretical and empirical results establish new ZO optimization methods as a practical and theoretically grounded approach for resource-constrained LLM adaptation. Our code is available at this https URL

As the dense deployment of access points (APs) in cell-free massive multiple-input multiple-output (CF-mMIMO) systems presents significant challenges, per-AP coverage can be expanded using large-scale antenna arrays (LAAs). However, this approach incurs high implementation costs and substantial fronthaul demands due to the need for dedicated RF chains for all antennas. To address these challenges, we propose a hybrid beamforming framework that integrates wave-domain beamforming via stacked intelligent metasurfaces (SIM) with conventional digital processing. By dynamically manipulating electromagnetic waves, SIM-equipped APs enhance beamforming gains while significantly reducing RF chain requirements. We formulate a joint optimization problem for digital and wave-domain beamforming along with fronthaul compression to maximize the weighted sum-rate for both uplink and downlink transmission under finite-capacity fronthaul constraints. Given the high dimensionality and non-convexity of the problem, we develop alternating optimization-based algorithms that iteratively optimize digital and wave-domain variables. Numerical results demonstrate that the proposed hybrid schemes outperform conventional hybrid schemes, that rely on randomly set wave-domain beamformers or restrict digital beamforming to simple power control. Moreover, the proposed scheme employing sufficiently deep SIMs achieves near fully-digital performance with fewer RF chains in the high signal-to-noise ratios regime.

The paper proposes a modular-based approach to constraint handling in process optimization and control. This is partly motivated by the recent interest in learning-based methods, e.g., within bioproduction, for which constraint handling under uncertainty is a challenge. The proposed constraint handler, called predictive filter, is combined with an adaptive constraint margin and a constraint violation cost monitor to minimize the cost of violating soft constraints due to model uncertainty and disturbances. The module can be combined with any controller and is based on minimally modifying the controller output, in a least squares sense, such that constraints are satisfied within the considered horizon. The proposed method is computationally efficient and suitable for real-time applications. The effectiveness of the method is illustrated through a realistic simulation case study of glycosylation constraint satisfaction in continuous monoclonal antibody biosimilar production using Chinese hamster ovary cells, for which the metabolic network model consists of 23 extracellular metabolites and 126 reactions.

We study Bayesian inverse problems with mixed noise, modeled as a combination of additive and multiplicative Gaussian components. While traditional inference methods often assume fixed or known noise characteristics, real-world applications, particularly in physics and chemistry, frequently involve noise with unknown and heterogeneous structure. Motivated by recent advances in flow-based generative modeling, we propose a novel inference framework based on conditional flow matching embedded within an Expectation-Maximization (EM) algorithm to jointly estimate posterior samplers and noise parameters. To enable high-dimensional inference and improve scalability, we use simulation-free ODE-based flow matching as the generative model in the E-step of the EM algorithm. We prove that, under suitable assumptions, the EM updates converge to the true noise parameters in the population limit of infinite observations. Our numerical results illustrate the effectiveness of combining EM inference with flow matching for mixed-noise Bayesian inverse problems.

The two-star random graph is the simplest exponential random graph model with nontrivial interactions between the graph edges. We propose a set of auxiliary variables that control the thermodynamic limit where the number of vertices N tends to infinity. Such `master variables' are usually highly desirable in treatments of `large N' statistical field theory problems. For the dense regime when a finite fraction of all possible edges are filled, this construction recovers the mean-field solution of Park and Newman, but with an explicit control over the 1/N corrections. We use this advantage to compute the first subleading correction to the Park-Newman result, which encodes the finite, nonextensive contribution to the free energy. For the sparse regime with a finite mean degree, we obtain a very compact derivation of the Annibale-Courtney solution, originally developed with the use of functional integrals, which is comfortably bypassed in our treatment.

This article introduces HYLU, a hybrid parallel LU factorization-based general-purpose solver designed for efficiently solving sparse linear systems (Ax=b) on multi-core shared-memory architectures. The key technical feature of HYLU is the integration of hybrid numerical kernels so that it can adapt to various sparsity patterns of coefficient matrices. Tests on 34 sparse matrices from SuiteSparse Matrix Collection reveal that HYLU outperforms Intel MKL PARDISO in the numerical factorization phase by geometric means of 1.95X (for one-time solving) and 2.40X (for repeated solving). HYLU can be downloaded from this https URL.

We consider the class of Davies quantum semigroups modelling thermalization for translation-invariant Calderbank-Shor-Steane (CSS) codes in D dimensions. We prove that conditions of Dobrushin-Shlosman-type on the quantum Gibbs state imply a modified logarithmic Sobolev inequality with a constant that is uniform in the system's size. This is accomplished by generalizing parts of the classical results on thermalization by Stroock, Zegarlinski, Martinelli, and Olivieri to the CSS quantum setting. The results in particular imply the rapid thermalization at any positive temperature of the toric code in 2D and the star part of the toric code in 3D, implying a rapid loss of stored quantum information for these models.

With the increasing adoption of Large Language Models (LLMs), more customization is needed to ensure privacy-preserving and safe generation. We address this objective from two critical aspects: unlearning of sensitive information and robustness to jail-breaking attacks. We investigate various constrained optimization formulations that address both aspects in a \emph{unified manner}, by finding the smallest possible interventions on LLM weights that either make a given vocabulary set unreachable or embed the LLM with robustness to tailored attacks by shifting part of the weights to a \emph{safer} region. Beyond unifying two key properties, this approach contrasts with previous work in that it doesn't require an oracle classifier that is typically not available or represents a computational overhead. Surprisingly, we find that the simplest point-wise constraint-based intervention we propose leads to better performance than max-min interventions, while having a lower computational cost. Comparison against state-of-the-art defense methods demonstrates superior performance of the proposed approach.

When to apply wavelet analysis to real-time temporal signals, where the future cannot be accessed, it is essential to base all the steps in the signal processing pipeline on computational mechanisms that are truly time-causal. This paper describes how a time-causal wavelet analysis can be performed based on concepts developed in the area of temporal scale-space theory, originating from a complete classification of temporal smoothing kernels that guarantee non-creation of new structures from finer to coarser temporal scale levels. By necessity, convolution with truncated exponential kernels in cascade constitutes the only permissable class of kernels, as well as their temporal derivatives as a natural complement to fulfil the admissibility conditions of wavelet representations. For a particular way of choosing the time constants in the resulting infinite convolution of truncated exponential kernels, to ensure temporal scale covariance and thus self-similarity over temporal scales, we describe how mother wavelets can be chosen as temporal derivatives of the resulting time-causal limit kernel. By developing connections between wavelet theory and scale-space theory, we characterize and quantify how the continuous scaling properties transfer to the discrete implementation, demonstrating how the proposed time-causal wavelet representation can reflect the duration of locally dominant temporal structures in the input signals. We propose that this notion of time-causal wavelet analysis could be a valuable tool for signal processing tasks, where streams of signals are to be processed in real time, specifically for signals that may contain local variations over a rich span of temporal scales, or more generally for analysing physical or biophysical temporal phenomena, where a fully time-causal analysis is called for to be physically realistic.

We prove that an effective temperature naturally emerges from the algorithmic structure of a regular universal Turing machine (UTM), without introducing any external physical parameter. In particular, the redundancy growth of the machine's wrapper language induces a Boltzmann--like exponential weighting over program lengths, yielding a canonical ensemble interpretation of algorithmic probability. This establishes a formal bridge between algorithmic information theory and statistical mechanics, in which the adopted UTM determines the intrinsic ``algorithmic temperature.'' We further show that this temperature approaches its maximal limit under the universal mixture (Solomonoff distribution), and discuss its epistemic meaning as the resolution level of an observer.

We introduce a general framework for constructing generative models using one-dimensional noising processes. Beyond diffusion processes, we outline examples that demonstrate the flexibility of our approach. Motivated by this, we propose a novel framework in which the 1D processes themselves are learnable, achieved by parameterizing the noise distribution through quantile functions that adapt to the data. Our construction integrates seamlessly with standard objectives, including Flow Matching and consistency models. Learning quantile-based noise naturally captures heavy tails and compact supports when present. Numerical experiments highlight both the flexibility and the effectiveness of our method.