In this paper, we present a new formalism, the Field Tensor Network Integral Logical Operator (FTNILO), to obtain the explicit equation that returns the minimum, maximum, and zeros of a multivariable injective function, and an algorithm for non-injective ones. This method extends the MeLoCoToN algorithm for inversion and optimization problems with continuous variables, by using Field Tensor Networks. The fundamentals of the method are the conversion of the problem of minimization of $N$ continuous variables into a problem of maximization of a dependent functional of a single variable. It can also be adapted to determine other properties, such as the zeros of any function. For this purpose, we use an extension of the imaginary time evolution, the new method of continuous signals, and partial or total integration, depending on the case. In addition, we show a direct way to recover both the tensor networks and the MeLoCoToN from this formalism. We show some examples of application, such as the Riemann hypothesis resolution. We provide an explicit integral equation that gives the solution of the Riemann hypothesis, being that if it results in a zero value, it is correct; otherwise, it is wrong. This algorithm requires no deep mathematical knowledge and is based on simple mathematical properties.

Human-machine collaboration often involves constrained optimization problems for decision-making processes. However, when the machine is a dynamical system with a continuously evolving state, infeasibility due to multiple conflicting constraints can lead to dangerous outcomes. In this work, we propose a heuristic-based method that resolves infeasibility at every time step by selectively disregarding a subset of soft constraints based on the past values of the Lagrange multipliers. Compared to existing approaches, our method requires the solution of a smaller optimization problem to determine feasibility, resulting in significantly faster computation. Through a series of simulations, we demonstrate that our algorithm achieves performance comparable to state-of-the-art methods while offering improved computational efficiency.

Lecture note topics: 1. Some tools from real and complex analysis, 2. Hilbert spaces, 3. Banach spaces, 4. Compact operators and their spectra, 5. Intermezzo: reproducing kernel Hilbert spaces, 6. Banach algebras ,7. Spectral theory of unitary, and of self-adjoint operators

A compatible associative algebra is a vector space endowed with two associative multiplication operations that satisfy a natural compatibility condition. In this paper, we investigate and classify compatible pairs of associative algebras of complex dimension less than four. Alongside these classifications, we systematically compute and analyze various algebraic invariants associated with them, including derivations, centroids, automorphism groups, quasi-centroids, Rota-Baxter operators, Nijenhuis operators, averaging operators, Reynolds operators, quasi-derivations, and generalized derivations.

Using the Burklund-Schlank-Yuan abstraction of ``algebraically closed" to ``Nullstellensatzian", we show that a $G$-Tambara functor is Nullstellensatzian if and only if it is the coinduction of an algebraically closed field (for any finite group $G$). As a consequence we deduce an equivalence between the $K$-theory spectrum of any Nullstellensatzian $G$-Tambara functor with the $K$ theory of some algebraically closed field.

We study certain weight functions on $[-1,1]$ which are particular cases of the general weights considered by Bernstein and Szeg\"o. These weights are numbered by two positive integers and when these integers tend to infinity, these weights approximate weight functions on $\mathbb{R}$ considered by Ismail and Valent. We also consider modifications of these weight functions by a continuous variable $a>0$. These ideas are then used to find finite analogs of some improper integrals first studied by Glaisher and Ramanujan.

For any odd prime $p$ and any imaginary quadratic field~$K$, the $p$-tower group $G_K$ associated to~$K$ is the Galois group over $K$ of the maximal unramified pro-$p$-extension of~$K$. This group comes with an action of a finite group $\{1,\sigma\}$ of order~$2$ induced by complex conjugation and is known to possess a number of other properties, making it a so-called Schur $\sigma$-group. Its maximal abelian quotient is naturally isomorphic to the $p$-primary part of the narrow ideal class group of~$\CO_K$, and the Cohen-Lenstra heuristic gives a probabilistic explanation for how often this group is isomorphic to a given finite abelian $p$-group. The present paper develops an analogue of this heuristic for the full group~$G_K$. It is based on a detailed analysis of general pro-$p$-groups with an action of $\{1,\sigma\}$, which we call $\sigma$-pro-$p$-groups. We construct a probability space whose underlying set consists of $\sigma$-isomorphism classes of weak Schur $\sigma$-groups and whose measure is constructed from the principle that the relations defining $G_K$ should be randomly distributed according to the Haar measure. We also compute the measures of certain basic subsets, the result being inversely proportional to the order of the $\sigma$-automorphism group of a certain finite $\sigma$-$p$-group, as has often been observed before. Finally, we show that the $\sigma$-isomorphism classes of weak Schur $\sigma$-groups for which each open subgroup has finite abelianization form a subset of measure~$1$.

We prove a formula for weighted sums of the first $n$ coefficients of mock modular forms of moderate growth and apply it to Hurwitz class numbers and coefficients of negative half integral weight Eisenstein series, which take the form of certain quadratic Dirichlet $L$-values. Our formula is a mock modular version of a Bessel-sum identity proved by Chandrasekharan and Narasimhan for Dirichlet series satisfying a functional equation. Our proof utilizes $L$-functions for mock modular Eisenstein series defined by Shankadhar and Singh.

For any odd prime $p$, the Galois group of the maximal unramified pro-$p$-extension of an imaginary quadratic field is a Schur $\sigma$-group. But Schur $\sigma$-groups can also be constructed and studied abstractly. We prove that if $p>3$, any Schur $\sigma$-group of Zassenhaus type $(3,3)$, for which every open subgroup has finite abelianization, is isomorphic to an open subgroup of a form of ${\rm PGL}_2$ over ${\mathbb Q}_p$. Combined with earlier work on an analogue of the Cohen-Lenstra heuristic for Schur $\sigma$-groups, or with the Fontaine-Mazur conjecture, this lends credence to one direction of the (3,3)-conjecture.

Static manifolds with boundary were recently introduced by Cruz and Vit\'orio in the context of the prescribed scalar curvature problem in a manifold with boundary with prescribed mean curvature. This kind of manifold is also interesting from the point of view of the general theory of relativity. In this article, we introduce electrostatic manifolds with boundary as a natural generalization of static manifolds with boundary in the presence of a non-zero electric field. We study the geometry of the zero-level set of the potential and its connection to the global properties of electrostatic manifolds with boundary. In particular, we establish some rigidity theorems for the 3-dimensional Euclidean ball and for the Reissner-Nordstr\"om manifold.

Consider an algebraic two-level method applied to the $n$-dimensional linear system $A \mathbf{x} = \mathbf{b}$ using fine-space preconditioner (i.e., ``relaxation'' or ``smoother'' in the context of multigrid) $M$, with $M \approx A$, restriction and interpolation $R$ and $P$, and algebraic coarse-space operator ${A_c := R^*AP}$. Then, what are the the best possible transfer operators $R$ and $P$ of a given dimension $n_c < n$? Brannick et al. (2018) showed that when $A$ and $M$ are Hermitian positive definite (HPD), the optimal interpolation is such that its range contains the $n_c$ smallest generalized eigenvectors of the matrix pencil $(A, M)$. Recently, Ali et al. (2025) generalized this framework to the non-HPD setting, by considering both right (interpolation) and left (restriction) generalized eigenvectors of $(A, M)$, but were unable to show norm-based convergence or optimality. Moreover, the transfer operators as derived therein are typically complex valued, which is not practical for real-valued problems, as usually arise in the context of numerical partial differential equations, for example. Here we significantly strengthen the results from Ali et al., first characterizing all inner products in which the resulting coarse-space correction defined by these transfer operators is orthogonal. We then develop tight two-level convergence bounds in these norms, and prove that the underlying transfer operators are genuinely optimal in these norms. As a special case, our theory both recovers and extends the HPD results from Brannick et al. Finally, we show how to construct optimal, real-valued transfer operators in the case of that $A$ and $M$ are real valued, but are not symmetric positive definite. Numerical examples arising from discretized advection and wave-equation problems are used to verify and illustrate the theory.

The celebrated Artin conjecture on primitive roots asserts that given any integer $g$ which is neither $-1$ nor a perfect square, there is an explicit constant $A(g)>0$ such that the number $\Pi(x;g)$ of primes $p\le x$ for which $g$ is a primitive root is asymptotically $A(g)\pi(x)$ as $x\to\infty$, where $\pi(x)$ counts the number of primes not exceeding $x$. Artin's conjecture has remained unsolved since its formulation 98 years ago. Nevertheless, Hooley demonstrated in 1967 that Artin's conjecture is a consequence of the Generalized Riemann Hypothesis (GRH) for Dedekind zeta functions of certain Kummer extensions over $\mathbb{Q}$. In this paper, we establish the Artin--Hooley asymptotic formula, under GRH, whenever $\log{x}/\log\log{2|g|} \to \infty$. Under GRH, we also show that the least prime $p_g$ possessing $g$ as a primitive root satisfies the upper bound $p_g=O(\log^{19}(2|g|))$ uniformly for all non-square $g\ne-1$. We conclude with an application to the average value of $p_g$ as well as discussion of an analogue concerning the least ``almost-primitive'' root $p_{g}^{\ast}$.

We establish a best-possible minimum codegree condition for the existence of a perfect tiling of a $3$-uniform hypergraph $H$ with copies of the generalised triangle $T$, which is the 3-uniform hypergraph with five vertices $a, b, c, d, e$ and three edges $abc$, $abd$, $cde$. We also give an asymptotically-optimal minimum codegree condition for the rainbow version of the problem.

We construct numerous fields with $m$-invariant not equal to $2^n$ for any integer $n \geq 0$.

In this article, we introduce balance equations over commutative rings $R$ and associate $R$-weighted graphs to them so that solving balance equations corresponds to a consistent labeling of vertices of the associated graph. Our primary focus is the case when $R$ is a commutative local ring whose residue field contains at least three elements. In this case, we provide explicit solutions of balance equations. As an application, letting $R$ to be the ring of $p$-adic integers, we examine some necessary and sufficient conditions for a $p$-group of nilpotency class $2$ to have its set of commutators coincide with its commutator subgroup. We also apply our results to study the surjectivity of the Lie bracket in Lie algebras, without any restriction on their dimension and the underlined field.

We prove that for volume preserving, partially hyperbolic, center bunched endomorphisms with constant Jacobian, essential accessibility implies ergodicity.

Divergence and vorticity damping are explicit diffusion mechanisms used in dynamical cores to ensure numerical stability. There is a mesh-dependent upper bound on the corresponding diffusion coefficient, else the diffusion itself becomes a source of instability. This work considers such stability limits for three gnomonic cubed-sphere meshes -- 1) equidistant, 2) equiangular, and 3) equi-edge mappings. Von Neumann analysis is used to derive linear stability limits, which show that the stability of divergence and vorticity damping depends on the cell areas and aspect ratios of the cubed-sphere grid. The linear theory is compared to practical divergence and vorticity damping limits in the CAM-FV3 dynamical core, using a baroclinic wave test case on the equiangular and equi-edge grids. For divergence damping, both the magnitude of maximum stable coefficients and the locations of instability agree with linear theory. Due to implicit vorticity diffusion from the transport scheme, practical limits for vorticity damping are lower than the explicit stability limits. The maximum allowable vorticity damping coefficient varies with the choice of horizontal transport scheme for the equi-edge grid; it is hypothesised that this indicates the relative implicit diffusion of the transport scheme in this test.

In 2002, Kamae and Zamboni introduced maximal pattern complexity and determined that any aperiodic sequence must have maximal pattern complexity at least $2k$. In 2006, Kamae and Rao examined the maximal pattern complexity of sequences over larger alphabets and showed that sequences which have maximal pattern complexity less than $\ell k$, for $\ell$ the size of the alphabet, must have some periodic structure. In this paper, we investigate the structure of sequences of low maximal pattern complexity over $\ell$ letters where $\liminf\limits\limits_{k \to \infty} p_{\alpha}^*(k) - 3k = -\infty$. In addition, we show that the minimal maximal pattern complexity of an aperiodic sequence which uses all $\ell$ letters is $p_{\alpha}^*(k) = 2k + \ell -2$, and give an exact structure for aperiodic sequences with this maximal pattern complexity.

A surface is considered flexible if it allows a continuous deformation that preserves both metric and smoothness. We introduce a novel construction method, called "base + crinkle," for generating a broad class of non-self-intersecting flexible closed polyhedral surfaces. These surfaces may be non-triangulated, exhibit multiple kinematic degrees of freedom, and possess topologies beyond the sphere. We further discuss the broader applicability and potential generalizations of the method, offering complementary insights into the geometry of origami and the design of engineering mechanisms.

We consider the problem of counting $k$-tuples of positive integers satisfying any arbitrary set of gcd conditions, where every integer is not larger than $x$. We first establish the conditions to guarantee the existence of such tuples, and then obtain asymptotic formulae for the count of such tuples with the help of a multivariable Dirichlet series. Part of this work can be viewed as a generalization of T\'oth's work, where the conditions are pairwise.

The aim of this paper is to study the stability of solutions to the general weighted system with negative exponents: \( \Delta_s u = \rho(\mathbf{x}) v^{-p}, \quad \Delta_s v = \rho(\mathbf{x}) u^{-\theta}, \quad u,v > 0 \) in \( \mathbb{R}^N \), where \( p \geq \theta > 1 \) and \( s \geq 0 \). Here, \( \Delta_s u = \Delta_x u + |x|^{2s} \Delta_y u \) is the Grushin operator, and \( \rho \) is a nonnegative continuous function satisfying certain conditions. We show that the system has no stable solution if \( p \geq \theta > 1 \) and \( N_s < 2 \left[ 1 + (2 + \alpha)x_0 \right] \), where \( x_0 \) is the largest root of the equation \( x^4 - \frac{16p\theta(p-1)}{\theta-1} \left( \frac{1}{p+\theta+2} \right)^2 \left[ x^2 + \frac{p+\theta-2}{(p+\theta+2)(\theta-1)} x + \frac{p-1}{(\theta-1)(p+\theta+2)^2} \right] = 0 \). Our result improves previous work and also applies to the weighted equation \( \Delta_s u = \rho(\mathbf{x}) u^{-p} \) in \( \mathbb{R}^N \), where \( p > 1 \).

Let $n \geq 2$ be an integer and let $K$ be a number field with ring of integers $\mathcal{O}_K$. We prove that the set of ternary $n$-ic forms with coefficients in $\mathcal{O}_K$ and fixed nonzero discriminant, breaks up into finitely many $\mathrm{GL}_3(\mathcal{O}_K)$-orbits. This generalizes a result of Birch--Merriman in the binary forms case. We also prove a similar finiteness result on the $\mathrm{GL}_3(\mathcal{O}_K)$-orbits of the 27-dimensional representation of $\mathrm{GL}_3$ with highest weight $(4, 2)$.

In this paper, we study certain Banach spaces of analytic functions on which a left-invertible multiplication operator acts. It turns out that, under natural conditions, its left inverse is a Cowen-Douglas operator. We investigate how the analytic continuations of functions from an invariant subspace of this Cowen-Douglas operator relate to the spectrum of its restriction to that subspace.

We show that cyclic permutations avoiding $321$ are precisely those permutations whose image under the fundamental bijection avoid a set of vincular patterns. We do this by using pattern functions and arrow patterns, in combination with the characterization of $321$ avoidance in terms of equality of the upper bound of the Daiconis-Graham inequalities. We then explore some consequences of this result, including upper and lower bound results on the growth rate of $321$ avoiding cycles.

We construct and analyze a projection-free linearly implicit method for the approximation of flows of harmonic maps into spheres. The proposed method is unconditionally energy stable and, under a sharp discrete regularity condition, achieves second order accuracy with respect to the constraint violation. Furthermore, the method accommodates variable step sizes to speed up the convergence to stationary points and to improve the accuracy of the numerical solutions near singularities, without affecting the unconditional energy stability and the constraint violation property. We illustrate the accuracy in approximating the unit-length constraint and the performance of the method through a series of numerical experiments, and compare it with the linearly implicit Euler and two-step BDF methods.

Chromatic-choosablility is a notion of fundamental importance in list coloring. A graph is chromatic-choosable when its chromatic number is equal to its list chromatic number. In 1990, Kostochka and Sidorenko introduced the list color function of a graph $G$, denoted $P_{\ell}(G,m)$, which is the list analogue of the chromatic polynomial of $G$, $P(G,m)$. It is known that for any graph $G$ there is a positive integer $k$ such that $P_{\ell}(G,m) = P(G,m)$ whenever $m \geq k$. In this paper, we study enumerative chromatic-choosability. A graph $G$ is enumeratively chromatic-choosable when $P_{\ell}(G,m) = P(G,m)$ whenever $m \in \mathbb{N}$. We completely determine the graphs of chromatic number two that are enumeratively chromatic-choosable. We construct examples of graphs that are chromatic-choosable but fail to be enumeratively-chromatic choosable, and finally, we explore a conjecture as to whether for every graph $G$, there is a $p \in \mathbb{N}$ such that the join of $G$ and $K_p$ is enumeratively chromatic-choosable. The techniques we use to prove results are diverse and include probabilistic ideas and ideas from DP (or correspondence)-coloring.

Constructions of regular heptagon and triskaidecagon by trisection of an angle are well known. An elegant construction of the heptagon by S. Adlaj shows a 3-fold symmetry related to a Galois group. Based on the later construction, in this article one more for the heptagon and two more for the triskaidecagon are presented, all using angle trisection.

Time Series Classification (TSC) is an important problem with numerous applications in science and technology. Dissimilarity-based approaches, such as Dynamic Time Warping (DTW), are classical methods for distinguishing time series when time deformations are confounding information. In this paper, starting from a deformation-based model for signal classes we define a problem statement for time series classification problem. We show that, under theoretically ideal conditions, a continuous version of classic 1NN-DTW method can solve the stated problem, even when only one training sample is available. In addition, we propose an alternative dissimilarity measure based on Optimal Transport and show that it can also solve the aforementioned problem statement at a significantly reduced computational cost. Finally, we demonstrate the application of the newly proposed approach in simulated and real time series classification data, showing the efficacy of the method.

We define the log-gamma sheet and the log-gamma landscape in terms of the 2-parameter and 4-parameter free energy of the log-gamma polymer model and prove that they converge to the Airy sheet and the directed landscape, which are central objects in the Kardar-Parisi-Zhang (KPZ) universality class. Our proof of convergence to the Airy sheet relies on the invariance of free energy through the geometric RSK correspondence and the monotonicity of the free energy. To upgrade the convergence to the directed landscape, tail bounds in both spatial and temporal directions are required. However, due to the lack of scaling invariance in the discrete log-gamma polymer--unlike the Brownian setting of the O'Connell-Yor model--existing on-diagonal fluctuation bounds are insufficient. We therefore develop new off-diagonal local fluctuation estimates for the log-gamma polymer.

We prove infinitely many cases of conjectured sharp upper and lower bounds for the spanning tree entropy of any planar lattice graph. These bounds come from volumes of associated hyperbolic alternating links, right-angled hyperbolic polyhedra and hyperbolic regular ideal bipyramids. For many planar lattice graphs, we show these bounds are easy to compute and provide excellent numerical estimates for the spanning tree entropy.

In this paper we introduce the conformal fractional Dirac operator and its associated fractional spinorial Yamabe problem. We also present a Caffarelli-Silvestre type extension for this fractional operator, allowing us to express it as a Dirichlet-to-Neumann type operator. As a consequence, we exhibit energy inequalities associated to this operator along with a weighted type Sobolev inequality for spinors. In the second part of the paper, we focus on the critical operator (which can be local or non-local depending on the evenness of the dimension). We introduce a Q-curvature operator, acting on spinors generalizing the classical notion of the scalar Q-curvature.

A Kakeya set in $\mathbb{R}^n$ is a compact set that contains a unit line segment $I_e$ in each direction $e \in S^{n-1}$. The Kakeya conjecture states that any Kakeya set in $\mathbb{R}^n$ has Hausdorff dimension $n$. We consider a restricted case where the midpoint of each line segment $I_e$ must belong to a fixed set $A$ with packing dimension at most $s \in [0, n]$. In this case, we show that the Hausdorff dimension of the Kakeya set is at least $n - s$. Furthermore, using the "bush argument", we improve the lower bound to $\max \{ n - s, n - g_n(s)\}$, where $g_n(s)$ is defined inductively. For example, when $n = 4$, we prove that the Hausdorff dimension is at least $\max\{\frac{19}{5} - \frac{3}{5}s,4-s\}$. We also establish Kakeya maximal function analogues of these results.

Semantic communication, due to its focus on the transmitting meaning rather than the raw bit data, poses unique security challenges compared to the traditional communication systems. In particular, semantic communication systems are vulnerable to the malicious attacks that focus on the semantic layer, with the intention of understanding or distorting the intended meaning of the transmitted privacy data. Diffusion models, a class of generative artificial intelligence (GenAI), are well-suited for ensuring data security to attack. Through iteratively adding and then removing noise, diffusion models can generate meaningful information despite the presence of the unknown noise. This article proposes a diffusion-based framework to enhance the security of semantic transmission for the attacks including eavesdropping and jamming. Specifically, the proposed framework incorporates both the artificial noise and natural channel noise into the forward process of the diffusion models during the semantic transmission, with the reverse process used to remove noise at the legitimate receiver. In the eavesdropping scenarios, the artificial noise is the friendly noise designed to prevent semantic eavesdropping. In the jamming scenarios, the artificial noise is the malicious jamming generated by the jammer, which disrupts the semantic transmission. The case studies show that the proposed diffusion-based framework is promising in securing the semantic transmission. We also consolidate several broad research directions associated with the proposed framework.

In this paper, we apply the power-partible reduction to study arithmetic properties of sums involving Delannoy numbers $D_k$ and polynomials $D_k(z)$. Let $v\in\bN$ and $p$ be an odd prime. It is proved that, for any $z\in\bZ\setminus\{0,-1\}$, there exist $c_v\in z^{-v}\bZ[z]$ and $\tilde{c}_v\in (z+1)^{-v}\bZ[z]$, both free of $p$ and can be determined mechanically, such that \begin{equation*} \sum_{k=0}^{p-1}(2k+1)^{2v}D_k(z)\equiv c_v \left(\frac{-z}{p}\right) \pmod {p} \end{equation*} if $\gcd(p,z)=1$ and \begin{equation*} \sum_{k=0}^{p-1}(-1)^k(2k+1)^{2v}D_k(z)\equiv \tilde{c}_v \left(\frac{z+1}{p}\right) \pmod {p} \end{equation*} if $\gcd(p,z+1)=1$. Here $(-)$ denotes the Legendre symbol. When $n$ is a power of $2$, we find there exist odd integers $\rho_v$ and even integers $\tilde{\rho}_v$, both independent of $n$ and can be determined mechanically, such that \[ \sum_{k=0}^{n-1}(2k+1)^{2v+1}D_k\equiv \rho_v n \pmod {n^3} \] and \[ \sum_{k=0}^{n-1}(-1)^k(2k+1)^{2v+1}D_k\equiv \tilde{\rho}_v n^2 \pmod {n^3}. \] The case $v=1$ in the last congruence confirms a conjecture of Guo and Zeng in 2012.

In this paper, we investigate the global well-posedness and scattering theory for the defocusing energy supcritical inhomogeneous nonlinear Schr\"odinger equation $iu_t + \Delta u =|x|^{-b} |u|^\alpha u$ in four space dimension, where $s_c := 2- \frac{2-b}{\alpha} \in (1, 2)$ and $0<b<\min \{ (s_c-1)^2+1,3-s_c\}$. We prove that if the solution has a prior bound in the critical Sobolev space, that is, $u \in L_t^\infty(I; \dot{H}_x^{s_c}(\mathbb{R}^4))$, then $u$ is global and scatters. The proof of the main results is based on the concentration-compactness/rigidity framework developed by Kenig and Merle [Invent. Math. 166 (2006)], together with a long-time Strichartz estimate, a spatially localized Morawetz estimate, and a frequency-localized Morawetz estimate.

In this paper, we estimate the number of $\mathbb{F}_q$-primitive points on the affine hypersurface defined by the equation $f(x_1,\ldots,x_s)=0$, where $f\in\mathbb{F}_q[x_1,\dots,x_s]$ is an appropriate polynomial. In particular, we provide existence results for the case when $f$ is Dwork-regular and when $f$ is of Fermat type. Additionally, we present a proof for a recently posed conjecture. Finally, in the case where $q$ is a Fermat prime, we provide an explicit formula for the number of $\mathbb{F}_q$-primitive points on hyperplanes.

Given that $a,b\in\mathbb N$, $c_0,c_1\in\mathbb Z$, $(c_0,c_1)\neq (0,0)$, and a generalized Fibonacci sequence $(s_n)_{n\geq 0}$ where $s_0 = c_0$, $s_1 = c_1$, and $s_{n+1}=as_{n}+bs_{n-1}$ for all positive integers $n$. In this paper, we get the result that for every polynomials $P(x)$ with real coefficients, we can always find three polynomials $F_1(x), G_1(x), H_1(x)$ (not necessarily distinct) with real coefficients satisfying the identity: $\;2\sum_{k=1}^{n}P(k)s_{k-1} = F_1(n)s_{n+1} + G_1(n)s_n + H_1(n), \;\forall n\in\mathbb N$. Furthermore, we serve two constraints for $(s_n)_{n\geq 0}$: one constraint implies that there are infinitely many triples $(F_1(x), G_1(x), H_1(x))$ satisfying the identity $\;2\sum_{k=1}^{n}P(k)s_{k-1} = F_1(n)s_{n+1} + G_1(n)s_n + H_1(n), \;\forall n\in\mathbb N$, while another constraint implies that there is only one triple $(F_1(x), G_1(x), H_1(x))$ satisfying the identity $\;2\sum_{k=1}^{n}P(k)s_{k-1} = F_1(n)s_{n+1} + G_1(n)s_n + H_1(n), \;\forall n\in\mathbb N$.

In this paper, we delve into the dynamical properties of a class of three-dimensional logistic ecological models. By using the complete discriminant theory of polynomials, we first give a topological classification for each fixed point and investigate the stability of corresponding system near the fixed points. Then employing the bifurcation and normal form theory, we discuss all possible codimension-1 bifurcations near the fixed points, i.e., transcritical, flip, and Neimark-Sacker bifurcations, and further prove that the system can undergo codimension-2 bifurcations, specifically 1:2, 1:3, 1:4 strong resonances and weak resonance Arnold tongues. Additionally, chaotic behaviors in the sense of Marotto are rigorously analyzed. Numerical simulations are conducted to validate the theoretical findings and illustrate the complex dynamical phenomena identified.

We characterize all algebraic subsets of the tridisk that are Caratheodory sets, that is the intrinsic Caratheodory metric on the set equals the Caratheodory metric for the tridisk. We show that such sets are either retracts, or are isomorphic to one particular exceptional set.

For any given Cuntz semigroup, we introduce a function associated to it, called the ``Rank Ratio Function". This function ensures a global control for the oscillation of the rank of any pair of elements in the Cuntz semigroup and, also, plays a very important role in computing the relative radii of comparison of the Cuntz semigroups. Along the way, as we study both the C*-algebraic and algebraic aspects of the relative radius of comparison with a deeper analysis, we introduce another function called the ``Radius of Comparison Function" which provides a method for manufacturing non-classifiable C*-algebras with different relative radii of comparison.

Complex spectra of transfer and Koopman operators describe rotational motion in dynamical systems. A particularly relevant situation in applications is where the rotation speed depends on the position in the phase space. We consider a canonical model of such dynamics in the presence of small noise, and provide precise characterisations of the eigenspectrum and eigenfunctions of the corresponding transfer operators. Further, we study the limiting behaviour of the eigenspectrum and eigenfunctions in the zero-noise limit, including their quadratic and linear response. Our results clarify the structure of transfer and Koopman operator eigenspectra, and provide new interpretations for applications. Our theorems on support localisation of the eigenfunctions yield simple algorithms to detect the existence and phase-space location of approximately cyclic motion with distinct periods. Our response results demonstrate that information on the cycle periods and their locations determined by the operator eigendata is largely insensitive to the noise level. We believe the mechanisms creating the eigendata apply more broadly and enhance our understanding of approximate cycle detection in dynamical systems with operator methods.

Let $M$ be a hyperbolic Riemann surface with the first eigenvalue $\lambda_1(M)>0$. Let $\rho$ denote the distance from a fixed point $x_0\in{M}$ and $r_x$ the injectivity radius at $x$. We show that there exists a numerical constant $c_0>0$ such that if $r_x\ge c_0 \lambda_1(M)^{-3/4} \rho(x)^{-1/2}$ holds outside some compact set of $M$, then the Bergman distance verifies $d_B(x,x_0) \gtrsim \log [1+\rho(x)]$.

Quantization of field-theoretic models with gauge symmetries is often obstructed by quantum anomalies. It is commonly believed that the origin of these anomalies lies in the infinite number of degrees of freedom, which requires completing the model within an appropriate regularization scheme. This paper provides an explicit example of a finite-dimensional Hamiltonian system with first-class constraints whose quantization exhibits anomalies. These anomalies arise from the nontrivial topology of the reduced phase space.

In this article, we develop a framework for the joint functional calculus of commuting tuples of $\text{Ritt}_{\text{E}}$ operators on Banach spaces. We establish a transfer principle that relates the bounded holomorphic functional calculus for tuples of $\text{Ritt}_{\text{E}}$ operators to that of their associated sectorial counterparts. In addition, we prove a joint dilation theorem for commuting tuples of $\text{Ritt}_{\text{E}}$ operators on a broad class of Banach spaces. As a key application, we obtain an equivalent set of criteria on $L^p$-spaces for $1<p< \infty$ that determine when a commuting tuple of $\text{Ritt}_{\text{E}}$ operators admits a joint bounded functional calculus.

In this paper we establish a stability barrier of a class of high-order Hermite-type discretization of 1D advection equations underlying the hybrid-variable (HV) and active flux (AF) methods. These methods seek numerical approximations to both cell-averages and nodal solutions and evolves them in time simultaneously. It was shown in earlier work that the HV methods are supraconvergent, providing that the discretization uses more unknowns in the upwind direction than the downwind one, similar to the "upwind condition" of classical finite-difference schemes. Although it is well known that the stencil of finite-difference methods could not be too biased towards the upwind direction for stability consideration, known as "stability barrier", such a barrier has not been established for Hermite-type methods. In this work, we first show by numerical evidence that a similar barrier exists for HV methods and make a conjecture on the sharp bound on the stencil. Next, we prove the existence of stability barrier by showing that the semi-discretized HV methods are unstable given a stencil sufficiently biased towards the upwind direction. Tighter barriers are then proved using combinatorical tools, and finally we extend the analysis to studying other Hermite-type methods built on approximating nodal solutions and derivatives, such as those widely used in Hermite WENO methods.

We prove sharp anti-concentration results for log-concave random variables on the real line in both the discrete and continuous setting. Our approach is elementary and uses majorization techniques to recover and extend some recent and not so recent results.

Given a graph $G$, the $k$-coloring graph $\mathcal{C}_k(G)$ is constructed by selecting proper $k$-colorings of $G$ as vertices, with an edge between two colorings if they differ in the color of exactly one vertex. The number of vertices in $\mathcal{C}_k(G)$ is the famous chromatic polynomial of $G$. Asgarli, Krehbiel, Levinson and Russell showed that for any subgraph $H$, the number of induced copies of $H$ in $\mathcal{C}_k(G)$ is a polynomial function in $k$. Hogan, Scott, Tamitegama, and Tan found a shorter proof for polynomiality of these chromatic $H$-polynomials. In this paper, we provide a method of constructing these polynomials explicitly in terms of chromatic polynomials of shadow graphs. We illustrate the practicality of our formulas by computing an explicit formula for $H$-polynomial for trees when $H=Q_d$ is an arbitrary hypercube, a task which does not seem approachable from previous methods. The coefficients of the resulting polynomials feature generalized degree sequences introduced by Crew. In the special case when $H=P_2$, the corresponding polynomial is dubbed the chromatic pairs polynomial. We present a pair of graphs $G_1$ and $G_2$ sharing the same chromatic pairs polynomial but different chromatic polynomials, disproving a conjecture raised by Asgarli, Krehbiel, Levinson and Russell.

Zhang et al.(2020) considered the visibility of self-similar set and gave a complete description of visible set. In this paper, we further observe the visibility problems on non-self-similar sets, in this case the structure becomes more complicated.

The Hermitian Veronesean in $PG(3,q^2)$, given by $\mathcal{V}:=\{ (1,x,x^q,x^{q+1}):x\in\mathbb{F}_q\}\cup\{(0,0,0,1)\}$, is a well-studied rational curve, and forms a {\em special} set of the Hermitian surface $H(3,q^2)$. In this paper, we give two local characterisations of the Hermitian Veronesean, based on sublines and triples of points in perspective.

We study a fuzzy Boltzmann equation, where collisions are delocalised and modulated by a spatial kernel. We show that as the spatial kernel converges to a delta distribution, the solutions to these equations converge to renormalised solutions of the inhomogeneous Boltzmann equations.

We define non-pivotal analogs of the Temperley-Lieb, Motzkin, and planar rook monoids, and compute bounds for the sizes of their nontrivial simple representations. From this, we assess the two types of monoids in their relative suitability for use in cryptography by comparing their representation gaps and gap ratios. We conclude that the non-pivotal monoids are generally worse for cryptographic purposes.

We begin by defining general hypergeometric functions over finite fields and proving a finite field analogue of a classical symmetry in their complex counterparts. Next, we up grade this symmetry to explicit isomorphisms between algebraic varieties defined by products of Fermat hypersurfaces and Artin-Schreier curves. The numbers of rational points on these varieties are hypergeometric functions over finite fields.

We consider the Dirac equations in polar form proving that they can equivalently be re-configured into a system of equations consisting of derivatives of the velocity density plus the Hamilton-Jacobi equation, giving the momentum in terms of relativistic quantum potentials (i.e. displaying first-order derivatives of the two degrees of freedom of the spinor field): this system is said to have Madelung structure. Conservation laws, second-order equations and multi-valuedness are also discussed.

Baxter numbers are known as the enumeration of Baxter permutations and numerous other discrete structures, playing a significant role across combinatorics, algebra, and analysis. In this paper, we focus on the analytic properties related to Baxter numbers. We prove that the descent polynomials of Baxter permutations have interlacing zeros, which is a property stronger than real-rootedness. Our approach is based on Dilks' framework of $(q,t)$-Hoggatt sums, which is a $q$-analog for Baxter permutations. Within this framework, we show that the family of $(1,t)$-Hoggatt sums satisfies the interlacing property using fundamental results on Hadamard products of polynomials. For Baxter numbers, we prove their asymptotic $r$-log-convexity via asymptotic expansions of $P$-recursive sequences. In particular, we confirm their $2$-log-convexity using symbolic computation techniques.

One of the very few mathematically rigorous nonlinear model reduction methods is the restriction of a dynamical system to a low-dimensional, sufficiently smooth, attracting invariant manifold. Such manifolds are usually found using local polynomial approximations and, hence, are limited by the unknown domains of convergence of their Taylor expansions. To address this limitation, we extend local expansions for invariant manifolds via Pad\'e approximants, which re-express the Taylor expansions as rational functions for broader utility. This approach significantly expands the range of applicability of manifold-reduced models, enabling reduced modeling of global phenomena, such as large-scale oscillations and chaotic attractors of finite element models. We illustrate the power of globalized manifold-based model reduction on several equation-driven and data-driven examples from solid mechanics and fluid mechanics.

In mixed-autonomy traffic networks, autonomous vehicles (AVs) are required to make sequential routing decisions under uncertainty caused by dynamic and heterogeneous interactions with human-driven vehicles (HDVs). Early-stage greedy decisions made by AVs during interactions with the environment often result in insufficient exploration, leading to failures in discovering globally optimal strategies. The exploration-exploitation balancing mechanism inherent in multi-armed bandit (MAB) methods is well-suited for addressing such problems. Based on the Real-Time Dynamic Programming (RTDP) framework, we introduce the Upper Confidence Bound (UCB) exploration strategy from the MAB paradigm and propose a novel algorithm. We establish the path-level regret upper bound under the RTDP framework, which guarantees the worst-case convergence of the proposed algorithm. Extensive numerical experiments conducted on a real-world local road network in Shanghai demonstrate that the proposed algorithm effectively overcomes the failure of standard RTDP to converge to the optimal policy under highly stochastic environments. Moreover, compared to the standard Value Iteration (VI) framework, the RTDP-based framework demonstrates superior computational efficiency. Our results highlight the effectiveness of the proposed algorithm in routing within large-scale stochastic mixed-autonomy environments.

In this article, we consider perturbations of isometries on a compact Riemannian manifold $M$. We investigate the smooth (resp. analytic) rigidity phenomenon of groups of these isometries. As a particular case, we prove that if a finite family of smooth (resp. analytic) small enough perturbations is simultaneously conjugate to the family of isometries via a finitely smooth diffeomorphism, then it is simultaneously smoothly (resp. analytically) conjugate to it whenever the family of isometries satisfies a Diophantine condition. Our results generalize the rigidity theorems of Arnold, Herman, Yoccoz, Moser, etc. about circle diffeomorphisms which are small perturbations of rotations as well as Fisher-Margulis's theorem on group actions satisfying Kazhdan's property (T).

In [Bal18] a new method of constructing simplex designs based on cyclic group on $n$ elements has been proposed. One of the claims put forward therein is existence of 3-point simplex 3-design in dimension $d = 3$. In this manuscript we present explicit counterarguments and suggest a manner to rectify the existing proofs. By doing this, we show that the results presented in [Bal18] can be utilised to construct simplex 3-designs scaling as $d^2$, which suggest a general scaling of $d^{t-1}$. Finally, we put forward a notion that encompasses the objects conforming with bounds given in [Bal18], which we refer to as symmetry-restricted simplex $t$-designs.

We classify polar isometric actions on simply connected 3-dimensional Riemannian homogeneous spaces, up to orbit equivalence. In particular, we classify extrinsically homogeneous surfaces in such spaces and study the geometry of the orbit foliations of the corresponding cohomogeneity one actions.

We study the stability of solutions to a class of variational inequalities posed on obstacle-type convex sets, under Mosco-convergence. More precisely, for a fixed obstacle $\psi\in W_{0}^{1,p}(\Omega)\cap L^{\infty}(\Omega)$, we consider $u\in W_{0}^{1,p}(\Omega)\cap L^{\infty}(\Omega)$ satisfying $u\geq\psi$ a.e. and $$ \langle A(u),v-u\rangle+\int_{\Omega}H(x,u,\nabla u)(v-u)\geq 0$$ for all $v\in W_{0}^{1,p}(\Omega)\cap L^{\infty}(\Omega)$ with $v\geq\psi$. Here, $A$ is a Leray-Lions type operator, mapping $W_0^{1,p}(\Omega)$ into its dual $W^{-1, p'}(\Omega)$, while $H(x, u, D u)$ grows like $|D u|^p$. Our main result establishes that the solutions are stable under Mosco-convergence of the constraint sets. This extends classical stability results to natural growth problems.

The Modular Isomorphism Problem asks, if an isomorphism between modular group algebras of finite $p$-groups over a field $F$ implies an isomorphism of the group bases. We explore the differences of knowledge on the problem when $F$ is either assumed to be a prime field or a general field of characteristic $p$. After revising the literature and explaining reasons for the differences, we generalize some of the positive answers to the problem from the prime field case to the general case.

We establish $L^q$ spectral cluster bounds for families of orthonormal functions associated to the Laplace-Beltrami operator on a compact Riemannian manifold. The metric is only assumed to be of class $C^s$, where $0\leq s\leq 2$.

We investigate the relationship between symmetric functions and the representation theory of operads, relative operads, and PROPs. We extend the classical character map for symmetric sequences to relative bisymmetric sequences and symmetric bimodules. We introduce new operations on symmetric functions, the relative plethysm and the box product, which model via the character map the composition product of relative operads and the box product of PROPs. As applications, we include the computation of characters for stable twisted cohomology of automorphism groups of free groups and the Albanese cohomology of the IA-automorphism group.

Movable antennas (MAs) have recently garnered significant attention in wireless communications due to their capability to reshape wireless channels via local antenna movement within a confined region. However, to achieve accurate antenna movement, MA drivers introduce non-negligible mechanical power consumption, rendering energy efficiency (EE) optimization more critical compared to conventional fixed-position antenna (FPA) systems. To address this problem, we develop in this paper a fundamental power consumption model for stepper motor-driven MA systems by resorting to basic electric motor theory. Based on this model, we formulate an EE maximization problem by jointly optimizing an MA's position, moving speed, and transmit power. However, this problem is difficult to solve optimally due to the intricate relationship between the mechanical power consumption and the design variables. To tackle this issue, we first uncover a hidden monotonicity of the EE performance with respect to the MA's moving speed. Then, we apply the Dinkelbach algorithm to obtain the optimal transmit power in a semi-closed form for any given MA position, followed by an enumeration to determine the optimal MA position. Numerical results demonstrate that despite the additional mechanical power consumption, the MA system can outperform the conventional FPA system in terms of EE.

In this paper, we introduce a modified version of the renormalization group (RG) method and test its numerical accuracy. It has been tested on numerous scalar ODEs and systems of ODEs. Our method is primarily motivated by the possibility of simplifying amplitude equations. The key feature of our method is the introduction of a new homogeneous function at each order of the perturbation hierarchy, which is then used to remove terms from the amplitude equations. We have shown that there is a limit to how many terms can be removed, as doing so beyond a certain point would reintroduce linear growth. There is thus a \textit{core} in the amplitude equation, which consists of the terms that cannot be removed while avoiding linear growth. Using our modified RG method, higher accuracy can also be achieved while maintaining the same level of complexity in the amplitude equation.

In this paper, we study the asymptotic behavior of energy and action ground states to the following pseudo-relativistic Hartree equation \[ \left(\sqrt{-c^2\Delta +m^2c^4}-mc^2\right)u + \lambda u = \left(|x|^{-1}*|u|^2\right)u \] as the speed of light $c\to\infty$. We obtain an asymptotic expansion of the ground state as $c \to \infty,$ which is new in the case of the energy ground state and generalizes the results of Choi, Hong, and Seok (2018) for the action ground state.

For a generalized step reinforced random walk, starting from the origin, the first step is taken according to the first element of an innovation sequence. Then in subsequent epochs, it recalls a past epoch with probability proportional to a regularly varying sequence $\{\mu_n\}$ of index $\gamma>-1$; recalls and repeats the step taken with probability $p$, or with probability $1-p$ takes a fresh step from the innovation sequence. The innovation sequence is assumed to be i.i.d.\ with mean zero. We study the corresponding step reinforced random walk process with linearly scaled time as an r.c.l.l.\ function on $[0, \infty)$. We prove law of large numbers for the linearly scaled process almost surely and in $L^1$ for all possible values of $p$ and $\gamma$. Assuming finite second moments for the innovation sequence, we obtain interesting phase transitions based on the boundedness of a sequence associated with $\{\mu_n\}$. The random walk suitably scaled converges almost surely to a process, which may not be Gaussian, when the sequence is bounded and the convergence is in distribution to a Gaussian process otherwise. This phase transition introduces the point of criticality at $p_c=\frac{\gamma+1/2}{\gamma+1}$ for $\gamma>-\frac12$. For the subcritical regime, the process is diffusive, while it is superdiffusive otherwise. However, for the critical regime, the scaled process can converge almost surely or in distribution depending on the choice of sequence $\{\mu_n\}$. Almost sure convergence in the critical regime is new. In the critical regime, the scaling can include many more novel choices in addition to $\sqrt{n \log n}$. Further, we use linear time scale and time independent scales even for the critical regime. We argue the exponential time scale for the critical regime is not natural. All the convergences in all the regimes are obtained for the process as an r.c.l.l.\ function.

Since the fundamental work of Chow-Luo \cite{CL03}, Ge \cite{Ge12,Ge17} et al., the combinatorial curvature flow methods became a basic technique in the study of circle pattern theory. In this paper, we investigate the combinatorial Ricci flow with prescribed total geodesic curvatures in spherical background geometry. For infinite cellular decompositions, we establish the existence of a solution to the flow equation for all time. Furthermore, under an additional condition, we prove that the solution converges as time tends to infinity. To the best of our knowledge, this is the first study of an infinite combinatorial curvature flow in spherical background geometry.

We study list-recoverability of random linear codes over small fields, both from errors and from erasures. We consider codes of rate $\epsilon$-close to capacity, and aim to bound the dependence of the output list size $L$ on $\epsilon$, the input list size $\ell$, and the alphabet size $q$. Prior to our work, the best upper bound was $L = q^{O(\ell/\epsilon)}$ (Zyablov and Pinsker, Prob. Per. Inf. 1981). Previous work has identified cases in which linear codes provably perform worse than non-linear codes with respect to list-recovery. While there exist non-linear codes that achieve $L=O(\ell/\epsilon)$, we know that $L \ge \ell^{\Omega(1/\epsilon)}$ is necessary for list recovery from erasures over fields of small characteristic, and for list recovery from errors over large alphabets. We show that in other relevant regimes there is no significant price to pay for linearity, in the sense that we get the correct dependence on the gap-to-capacity $\epsilon$ and go beyond the Zyablov-Pinsker bound for the first time. Specifically, when $q$ is constant and $\epsilon$ approaches zero: - For list-recovery from erasures over prime fields, we show that $L \leq C_1/\epsilon$. By prior work, such a result cannot be obtained for low-characteristic fields. - For list-recovery from errors over arbitrary fields, we prove that $L \leq C_2/\epsilon$. Above, $C_1$ and $C_2$ depend on the decoding radius, input list size, and field size. We provide concrete bounds on the constants above, and the upper bounds on $L$ improve upon the Zyablov-Pinsker bound whenever $q\leq 2^{(1/\epsilon)^c}$ for some small universal constant $c>0$.

In this paper, we prove a Skoda type division theorem with sharp $L^2$-estimate. Furthermore, using this estimate, we provide new characterizations of plurisubharmonic functions. We also explain that the sharp $L^2$-division theorem leads the Guan-Zhou's sharp $L^2$-estimate for extension theorem.

Let $n\geq 2$ be an integer, $S_n$ be the Lie algebra of vector fields on $\mathbb{C}^n$ with zero divergence, and $D_n$ be the Weyl algebra over the polynomial algebra $A_n=\mathbb{C}[t_1,t_2,\cdots,t_n]$. In this paper, we study the simplicity of the tensor $S_n$-module $F(P,M)$, where $P$ is a simple $D_n$-module and $M$ is a simple $\mathfrak{sl}_n$-module. We obtain the necessary and sufficient conditions for $F(P,M)$ to be an irreducible module, and determine all simple subquotients of $F(P,M)$ when it is reducible.

In this paper, we analyze nonlinear model predictive control (MPC) using data-driven surrogates in the prediction and optimization step. First, we establish asymptotic stability of the origin, a controlled steady state, w.r.t. the MPC closed loop without stabilizing terminal conditions. To this end, we prove that cost controllability of the original system is preserved if proportional bounds on the approximation error hold. Here, proportional refers to state and control, while the respective constants depend on the approximation accuracy. The proportionality of the error bounds is a key element to derive asymptotic stability in presence of modeling errors and not only practical asymptotic stability. Second, we exemplarily verify the imposed assumptions for data-driven surrogates generated with kernel extended dynamic mode decomposition based on the Koopman operator. Hereby, we do not impose invariance assumptions on finite dictionaries, but rather derive all conditions under non-restrictive data requirements. Finally, we verify our findings with numerical simulations.

Computing geodesics for Riemannian manifolds is a difficult task that often relies on numerical approximations. However, these approximations tend to be either numerically unstable, have slow convergence, or scale poorly with manifold dimension and number of grid points. We introduce a new algorithm called GEORCE that computes geodesics via a transformation into a discrete control problem. We show that GEORCE has global convergence and quadratic local convergence. In addition, we show that it extends to Finsler manifolds. For both Finslerian and Riemannian manifolds, we thoroughly benchmark GEORCE against several alternative optimization algorithms and show empirically that it has a much faster and more accurate performance for a variety of manifolds, including key manifolds from information theory and manifolds that are learned using generative models.

This chapter reviews and compares discontinuous Galerkin time-stepping methods for the numerical approximation of second-order ordinary differential equations, particularly those stemming from space finite element discretization of wave propagation problems. Two formulations, tailored for second- and first-order systems of ordinary differential equations, are discussed within a generalized framework, assessing their stability, accuracy, and computational efficiency. Theoretical results are supported by various illustrative examples that validate the findings, enhancing the understanding and applicability of these methods in practical scenarios.

We extend Helgason's classical definition of a generalized Radon transform, defined for a pair of homogeneous spaces of an lcsc group $G$, to a broader setting in which one of the spaces is replaced by a possibly non-homogeneous dynamical system over $G$ together with a suitable cross section. This general framework encompasses many examples studied in the literature, including Siegel (or $\Theta$-) transforms and Marklof-Str\"ombergsson transforms in the geometry of numbers, Siegel-sVeech transforms for translation surfaces, and Zak transforms in time-frequency analysis. Our main applications concern dynamical systems $(X, \mu)$ in which the cross section is induced from a separated cross section. We establish criteria for the boundedness, integrability, and square-integrability of the associated Siegel-Radon transforms, and show how these transforms can be used to embed induced $G$-representations into $L^p(X, \mu)$ for appropriate values of $p$. These results apply in particular to hulls of approximate lattices and certain "thinnings" thereof, including arbitrary positive density subsets in the amenable case. In the special case of cut-and-project sets, we derive explicit formulas for the dual transforms, and in the special case of the Heisenberg group we provide isometric embedding of Schr\"odinger representations into the $L^2$-space of the hulls of positive density subsets of approximate lattices in the Heisenberg group by means of aperiodic Zak transforms.

The free positive multiplicative Brownian motion $(h_t)_{t\geq0}$ is the large $N$ limit in non-commutative distribution of matrix geometric Brownian motion. It can be constructed by setting $h_t:=g_{t/2}g_{t/2}^*$, where $(g_t)_{t\geq0}$ is a free multiplicative Brownian motion, which is the large $N$ limit in non-commutative distribution of the Brownian motion in $\operatorname{Gl}(N,\mathbb{C})$. One key property of $(h_t)_{t\geq0}$ is the fact that the corresponding spectral distributions $(\nu_t)_{t\geq0}\subset M^1((0,\infty))$ form a semigroup w.r.t. free multiplicative convolution. In recent work by M. Voit and the present author, it was shown that $\nu_t$ can be expressed by the image measure of a free additive convolution of the semicircle and the uniform distribution on an interval under the exponential map. In this paper, we provide a new proof of this result by calculating the moments of the free additive convolution of semicircle and uniform distributions on intervals. As a by-product, we also obtain new integral formulas for $\nu_t$ which generalize the corresponding known moment formulas involving Laguerre polynomials.

We thoroughly investigate Discontinuous Galerkin (DG) discretizations as time integrators for second-order oscillatory systems, considering both second-order and first-order formulations of the original problem. Key contributions include new convergence analyses for the second-order formulation and equivalence proofs between DG and classical time-stepping schemes (such as Newmark schemes and general linear methods). In addition, the chapter provides a detailed review and convergence analysis for the first-order formulation, alongside comparisons of the proposed schemes in terms of accuracy, consistency, and computational cost.

Minima selection is essential for defining Bi-Level Optimization (BLO) when the lower-level objective has multiple minimizers. While BLO is intractable in its general form, we show that restricting the lower-level objective to the recently proposed PL-circle functions (Gong et al., 2024) guarantees a continuous hyper-objective F_max. The PL-circle condition is strictly weaker than the global Polyak-Lojasiewicz condition used in prior BLO work and allows modeling of practical settings such as hyperparameter tuning in over-parameterized deep learning. However, even under this condition, F_max remains non-convex and non-smooth. To address this, we propose a relaxed solution concept: we approximate F_max with a continuously differentiable function F_max_tilde that is pointwise epsilon_v-close to F_max and seek an epsilon_g-stationary point of F_max_tilde. In this framework, we reduce the minima-selection subproblem to a sampling task using a novel Superquantile-Gibbs relaxation. By leveraging the manifold structure of the lower-level solution set under the PL-circle condition, our method finds a relaxed solution using poly(epsilon_v^{-k} epsilon_g^{-1}) queries to a Gibbs sampling oracle, which is efficiently implemented using Langevin dynamics. Here, k is the intrinsic dimension of the manifolds defining the lower-level solutions. This is the first work to characterize the complexity of BLO in terms of this intrinsic dimensionality.

A graph $G$ consists of two parts, the vertices and edges. The vertices constitute the vertex set $V(G)$ and the edges, the edge set. An edge \( e=xy \), \( ev \)-dominates not only the vertices incident to it but also those adjacent to either \( x \) or \( y \). The edge-vertex degree of $e,$ $deg^{ev}_{G}(e),$ is the number of vertices in the $ev$-dominating set of $e$. In this article, we compute expressions for the $ev$-degree version of the Zagreb index of several unary and binary graph operations.

This article uses Clifford algebra of definite signature to derive octonions and the Lie exceptional algebra G2 from calibrations using Pin(7). This is simpler than the usual exterior algebra derivation and uncovers a subalgebra of Spin(7) that enables G2 and an invertible element used to classify six other algebras which are found to be related to the symmetries of G2 in a way that breaks the symmetry of octonions. The 4-form calibration terms of Spin(7) are related to an ideal with three idempotents and provides a direct construction of G2 for each of the 480 representations of the octonions. Clifford algebra thus provides a new construction of G2 without using the Lie bracket. This result is extended to 15 dimensions generating another 100 algebras as well as the sedenions.

We prove that for $\delta$ small, $n$ large, and any three conjugacy classes $C_{1},C_{2},C_{3}$ of $G=\mathrm{Alt}(n)$ of size at least $|G|^{1-\delta}$ we have $C_{1}C_{2}C_{3}=G$. The result provides a positive answer to Problem 20.23 of the Kourovka Notebook [KM22], improves theorems of Garonzi and Mar\'oti [GM21] (using $4$ classes) and Rodgers [Rod02] (using larger classes), complements the known result for $G$ a simple group of Lie type [MP21] [LST24] [FM25], and is tight in several senses. Furthermore, since no character theory is involved, the proof can be used in principle to build a constructive algorithm that, given $g\in G$, outputs $c_{i}\in C_{i}$ such that $c_{1}c_{2}c_{3}=g$.

We identify isometric isomorphisms of the space of Kurzweil-Henstock integrable functions as bi-absolutely-continuous changes of variable.

We study the discretization of (almost-)Dirac structures using the notion of retraction and discretization maps on manifolds. Additionally, we apply the proposed discretization techniques to obtain numerical integrators for port-Hamiltonian systems and we discuss how to merge the discretization procedure and the constraint algorithm associated to systems of implicit differential equations.

Every graph is majority 3-choosable. This generalises the result by Shelah-Milner that every graph has an unfriendly 3-partition, confirming a conjecture of Haslegrave from 2020.

We introduce a new approach to the description of multi-sorted clones (sets of $k$-tuples of operations of the same arity, closed under coordinatewise composition and containing all projection tuples) on a two-element domain. Leveraging the well-known Galois connection between operations and relations, we define a small class of \textit{canonical relations} sufficient to describe all Boolean multi-sorted clones up to non-surjective operations. Furthermore, we introduce \textit{elementary operations} on relations, which are less cumbersome than general formulas and have many useful properties. Using these tools, we provide a~new and elementary proof of the famous Post's lattice theorem. We also show that every multi-sorted clone of $k$-tuples of operations decomposes into a surjective part described by canonical relations and $2k$ clones of $(k-1)$-tuples of operations. This structural understanding allows us to describe an embedding of the lattice of multi-sorted clones into a well-understood poset. In particular, we rederive -- by a simpler method -- a~result of V.~Taimanov originally from 1983, showing that every multi-sorted clone on a two-element domain is finitely generated. Finally, we also give a~concise proof of the Galois connection between (surjective) multi-sorted clones and the corresponding closed sets of relations.

In this paper, we study a class of inexact block triangular preconditioners for double saddle-point symmetric linear systems arising from the mixed finite element and mixed hybrid finite element discretization of Biot's poroelasticity equations. We develop a spectral analysis of the preconditioned matrix, showing that the complex eigenvalues lie in a circle of center $(1,0)$ and radius smaller than 1. In contrast, the real eigenvalues are described in terms of the roots of a third-degree polynomial with real coefficients. The results of numerical experiments are reported to show the quality of the theoretical bounds and illustrate the efficiency of the proposed preconditioners used with GMRES, especially in comparison with similar block diagonal preconditioning strategies along with the MINRES iteration.

We investigate the top of the spectrum of discrete Anderson Hamiltonians with correlated Gaussian noise in the large volume limit. The class of Gaussian noises under consideration allows for long-range correlations. We show that the largest eigenvalues converge to a Poisson point process and we obtain a very precise description of the associated eigenfunctions near their localisation centres. We also relate these localisation centres with the locations of the maxima of the noise. Actually, our analysis reveals that this relationship depends in a subtle way on the behaviour near $0$ of the covariance function of the noise: in some situations, the largest eigenfunctions are not associated with the largest values of the noise.

We study the geometry of compact strong HKT and, more generally, compact BHE manifolds. We prove that any compact BHE manifold with full holonomy must be K\"ahler and we establish a similar result for strong HKT manifolds. Additionally, we demonstrate a rigidity theorem for strong HKT structures on solvmanifolds and we completely classify those with parallel Bismut torsion. Finally, we introduce the Ricci foliation for hypercomplex manifolds and analyze its properties for compact, simply connected, 8-dimensional strong HKT manifolds, proving that they are always Hopf fibrations over a compact $4$-dimensional orbifold.

In previous work, categories of algebras of endofunctors were shown to be enriched in categories of coalgebras of the same endofunctor, and the extra structure of that enrichment was used to define a generalization of inductive data types. These generalized inductive data types are parametrized by a coalgebra $C$, so we call them $C$-inductive data types; we call the morphisms induced by their universal property $C$-inductive functions. We extend that work by incorporating natural transformations into the theory: given a suitable natural transformation between endofunctors, we show that this induces enriched functors between their categories of algebras which preserve $C$-inductive data types and $C$-inductive functions. Such $C$-inductive data types are often finite versions of the corresponding inductive data type, and we show how our framework can extend classical initial algebra semantics to these types. For instance, we show that our theory naturally produces partially inductive functions on lists, changes in list element types, and tree pruning functions.

We present a data-driven framework for learning dynamical systems on compact Riemannian manifolds based on the spectral exterior calculus (SEC). This approach represents vector fields as linear combinations of frame elements constructed using the eigenfunctions of the Laplacian on smooth functions, along with their gradients. Such reconstructed vector fields generate dynamical flows that consistently approximate the true system, while being compatible with the nonlinear geometry of the manifold. The data-driven implementation of this framework utilizes embedded data points and tangent vectors as training data, along with a graph-theoretic approximation of the Laplacian. In this paper, we prove the convergence of the SEC-based reconstruction in the limit of large data. Moreover, we illustrate the approach numerically with applications to dynamical systems on the unit circle and the 2-torus. In these examples, the reconstructed vector fields compare well with the true vector fields, in terms of both pointwise estimates and generation of orbits.

We give a relatively short and elementary proof of Manin's conjecture for split smooth quintic del Pezzo surfaces over the rational numbers.

Energy system optimization models are becoming increasingly popular for analyzing energy markets, such as the impact of new policies or interactions between energy carriers. One key challenge of these models is the trade-off between modeling accuracy and computational tractability. A recently proposed mathematical framework addresses this challenge by achieving exact time series aggregations merging time periods sharing the same active constraint sets. This aggregation, however, is insufficient when the number of unique active constraints is large. We overcome this issue by aggregating data points from different active constraint sets. While this further reduces model size, it inevitably introduces an error compared to the full model. Yet, we show how this error can be exactly quantified without re-solving the optimization problem, enabling users to trade off computational efficiency and model accuracy proactively. This may be especially useful in energy markets to accommodate varying granularity across short- and long-term time horizons.

In the setting where we have $n$ independent observations of a random variable $X$, we derive explicit error bounds in total variation distance when approximating the number of observations equal to the maximum of the sample (in the case where $X$ is discrete) or the number of observations within a given distance of an order statistic of the sample (in the case where $X$ is absolutely continuous). The logarithmic and Poisson distributions are used as approximations in the discrete case, with proofs which include the development of Stein's method for a logarithmic target distribution. In the absolutely continuous case our approximations are by the negative binomial distribution, and are established by considering negative binomial approximation for mixed binomials. The cases where $X$ is geometric, Gumbel and uniform are used as illustrative examples.

As observed by Joyal, the cohomology groups of the partition posets are naturally identified with the components of the operad encoding Lie algebras. This connection was explained in terms of operadic Koszul duality by Fresse, and later generalized by Vallette to the setting of decorated partitions. In this article, we set up and study a general formalism which produces a priori operadic structures (operads and operadic modules) on the cohomology of families of posets equipped with some natural recursive structure, that we call "operadic poset species". This framework goes beyond decorated partitions and operadic Koszul duality, and contains the metabelian Lie operad and Kontsevich's operad of trees as two simple instances. In forthcoming work, we will apply our results to the hypertree posets and their connections to post-Lie and pre-Lie algebras.

In this paper, we study the existence of normalized multi-bump solutions for the following Choquard equation \begin{equation*} -\epsilon^2\Delta u +\lambda u=\epsilon^{-(N-\mu)}\left(\int_{\mathbb{R}^N}\frac{Q(y)|u(y)|^p}{|x-y|^{\mu}}dy\right)Q(x)|u|^{p-2}u, \text{in}\ \mathbb{R}^N, \end{equation*} where $N\geq3$, $\mu\in (0,N)$, $\epsilon>0$ is a small parameter and $\lambda\in\mathbb{R}$ appears as a Lagrange multiplier. By developing a new variational approach, we show that the problem has a family of normalized multi-bump solutions focused on the isolated part of the local maximum of the potential $Q(x)$ for sufficiently small $\epsilon>0$. The asymptotic behavior of the solutions as $\epsilon\rightarrow0$ are also explored. It is worth noting that our results encompass the sublinear case $p<2$, which complements some of the previous works.

The Fourier Basis Density Model (FBM) was recently introduced as a flexible probability model for band-limited distributions, i.e. ones which are smooth in the sense of having a characteristic function with limited support around the origin. Its density and cumulative distribution functions can be efficiently evaluated and trained with stochastic optimization methods, which makes the model suitable for deep learning applications. However, the model lacked support for sampling. Here, we introduce a method inspired by discretization--interpolation methods common in Digital Signal Processing, which directly take advantage of the band-limited property. We review mathematical properties of the FBM, and prove quality bounds of the sampled distribution in terms of the total variation (TV) and Wasserstein--1 divergences from the model. These bounds can be used to inform the choice of hyperparameters to reach any desired sample quality. We discuss these results in comparison to a variety of other sampling techniques, highlighting tradeoffs between computational complexity and sampling quality.

A packing $k$-coloring of a graph $G$ is a partition of $V(G)$ into $k$ disjoint non-empty classes $V_1, \dots, V_k$, such that if $u,v \in V_i$, $i\in [k]$, $u\ne v$, then the distance between $u$ and $v$ is greater than $i$. The packing chromatic number of $G$ is the smallest integer $k$ which admits a packing $k$-coloring of $G$. In this paper, the packing chromatic number of the unitary Cayley graph of $\mathbb{Z}_n$ is computed. Two metaheuristic algorithms for calculating the packing chromatic number are also proposed.

The plethysm product of Schur functions corresponds to composition of polynomial representations of infinite general linear groups. Finding the plethysm coefficients $\langle s_\nu \circ s_\mu, s_\lambda\rangle$ that express an arbitrary plethysm $s_\nu \circ s_\mu$ as a sum $\sum_\lambda \langle s_\nu \circ s_\mu, s_\lambda \rangle s_\lambda$ of Schur functions is a fundamental open problem in algebraic combinatorics. We prove two stability theorems for plethysm coefficients under the operations of adding and/or joining an arbitrary partition to either $\mu$ or $\nu$. In both theorems $\mu$ may be replaced with an arbitrary skew partition. As special cases we obtain all stability results on plethysms of Schur functions in the literature to date. The proofs are entirely combinatorial using plethystic semistandard tableaux with positive and negative entries.

We derive an implicit numerical scheme for the solution of advection equation where the roles of space and time variables are exchanged using the inverse Lax-Wendroff procedure. The scheme contains a linear weight for which it is always second order accurate in time and space, and the stencil in the implicit part is fully upwinded for any value of the weight, enabling a direct computation of numerical solutions by forward substitution. To fulfill the local bounds for the solution represented by the discrete minimum and maximum principle (DMP), we use a predicted value obtained with the linear weight and check a priori if the DMP is valid. If not, we can use either a nonlinear weight or a limiter function that depends on Courant number and apply such a high-resolution version of the scheme to obtain a corrected value. The advantage of the scheme obtained with the inverse Lax-Wendroff procedure is that only in the case of too small Courant numbers, the limiting is towards the first order accurate scheme, which is not a situation occurring in numerical simulations with implicit schemes very often. In summary, the local bounds are satisfied up to rounding errors unconditionally for any Courant numbers, and the formulas for the predictor and the corrector are explicit. The high-resolution scheme can be extended straightforwardly for advection with nonlinear retardation coefficient with numerical solutions satisfying the DMP, and a scalar nonlinear algebraic equation has to be solved to obtain each predicted and corrected value. In numerical experiments, including transport on a sewer network, we can confirm the advantageous properties of numerical solutions for several representative examples.

We study smooth function spaces of Gelfand-Shilov type, with global behavior governed through a translation-invariant Banach function space and localized via a weight function system. We clarify the roles of the translation-invariant Banach function space, convolution, and pointwise multiplication in connection with the weight function system. Our primary goal is to characterize these function spaces - as well as the corresponding convolutor and multiplier spaces - through mollification. For this purpose, we introduce the moment-wise decomposition factorization property for pairs of compactly supported smooth functions, and establish complete characterizations in terms of mollifications with these windows.

Given elements $x,u,z$ in a finite group $G$ such that $z$ is the commutator of $x$ and $u$, and the orders of $x$ and $z$ divide respectively integers $k,m \geq 2$, and given an integer $r$ that is coprime to $k$ and $m$, there exists $w \in G$ such that the commutator of $x^r$ and $w$ is conjugate to $z^r$. If instead we are given elements $x,y,z \in G$ such that $xy = z$, whose respective orders divide integers $k,l,m \geq 2$, and are given an integer $r$ that is coprime to $k,l$ and $m$, then there exist $x'$, $y'$ and $z'$ conjugate to respectively $x^r$, $y^r$ and $z^r$ such that $x'y' = z'$. In this paper we completely answer the natural question for which values of $k,l,m,r$ every group has these properties. The proof uses combinatorial group theory and properties of the projective special linear group $\mathrm{PSL}_2(\mathbb{R})$.

We extend the following result of Cochran ``A closed $m$-quasi Einstein manifold ($M,g,X$) with $m \ne -2$ has constant scalar curvature if and only if $X$ is Killing" covering the missing accidental case $m=-2$ and generalize it showing that $X$ is Killing if the integral of the Lie derivative of the scalar curvature along $X$ is non-positive. For a closed $m$-quasi Einstein manifold of dimension $n$, if $n =2$, $X$ vanishes, and for $n>2$ if $X$ is conformal, then it is Killing; and in addition, if $M$ admits a non-Killing conformal vector field $V$, then it is globally isometric to a sphere and $V$ is gradient. Finally, we derive an integral identity for a vector field on a closed Riemannian manifold which provides a direct proof of the Bourguignon-Ezin conservation identity.

We consider the dynamical system with boundary control for the vector Schr\"odinger equation on the interval with a non-self-adjoint matrix potential. For this system, we study the inverse problem of recovering the matrix potential from the dynamical Dirichlet--to--Neumann operator. We first provide a method to recover spectral data for an abstract system from dynamic data and apply it to the Schr\"odinger equation. We then develop a strategy for solving the inverse problem for the Schr\"odinger equation using this method with other techniques of the Boundary control method.

We consider dynamic inverse problems for a dynamical system associated with a finite Jacobi matrix and for a system describing propagation of waves in a finite Krein-Stieltjes string. We offer three methods of recovering unknown parameters: entries of a Jacobi matrix in the first problem and point masses and distances between them in the second, from dynamic Dirichlet-to-Neumann operators. We also answer a question on a characterization of dynamic inverse data for these two problems.

We provide a categorification of Oh and Suh's combinatorial Auslander-Reiten quivers in the simply laced case. We work within the perfectly valued derived category $\mathrm{pvd}(\Pi_Q)$ of the 2-dimensional Ginzburg dg algebra of a Dynkin quiver $Q$. For any commutation class $[i]$ of reduced words in the corresponding Weyl group, we define a subcategory $C([i])$ of $\mathrm{pvd}(\Pi_Q)$ whose objects are obtained by applying a sequence of spherical twist functors to the simple objects. We describe the Hom-order for $C([i])$ in terms of $[i]$, generalizing a result of B\'edard. Furthermore, when $[i]$ is a commutation class for the longest element, we construct a category $D([i])$ generalizing the bounded derived category of $Q$. It is realized as a certain subquotient of $\mathrm{pvd}(\Pi_Q)$. We demonstrate the existence of particular distinguished triangles in $\mathrm{pvd}(\Pi_Q)$ with corners in $D([i])$, which allows us to extend the classical mesh-additivity to arbitrary commutation classes. Additionally, we define an analog of the Euler form and prove that its symmetrization yields the corresponding Cartan-Killing form. For commutation classes $[i]$ arising from Q-data, a generalization of Dynkin quivers with a height function introduced by Fujita and Oh, we establish the existence of a partially Serre functor on $D([i])$. Lastly, we apply our results to reinterpret a formula by Fujita and Oh for the inverse of the quantum Cartan matrix.

We consider a stationary variational inequality with gradient constraint and obstacle. We prove that this problem can be described by an equation using a Lagrange multiplier and a characteristic function. The Lagrange multiplier contains information about the contact set of the modulus of the gradient of the solution with the gradient constraint, and the characteristic function is defined in the contact set of the solution with the obstacle. Moreover, given a convergent sequence of data, we prove the stability of the corresponding solutions.

The representation theory of tensor functions is essential to constitutive modeling of materials including both mechanical and physical behaviors. Generally, material symmetry is incorporated in the tensor functions through a structural or anisotropic tensor that characterizes the corresponding point group. The general mathematical framework was well-established in the 1990s. Nevertheless, the traditional theory suffers from a grand challenge that many point groups involve fourth or sixth order structural tensors that hinder its practical applications in engineering. Recently, researchers have reformulated the representation theory and opened up opportunities to model anisotropic materials using low-order (i.e., 2nd-order and lower) structural tensors only, although the theory was not fully established. This work aims to fully establish the reformulated representation theory of tensor functions for all two-dimensional point groups. It was found that each point group needs a structural tensor set to characterize the symmetry. For each two-dimensional point group, the structural tensor set is proposed and the general tensor functions are derived. Only low-order structural tensors are introduced so researchers can readily apply these tensor functions for their modeling applications. The theory presented here is useful for constitutive modeling of materials in general, especially for composites, nanomaterials, soft tissues, etc.

In this paper, we propose an inexact Aleksandrov solution based Abedin-Kitagawa iteration (AKI) method for solving (real) Monge-Amp{\`e}re eigenvalue problems. The proposed approach utilizes the convergent Rayleigh inverse iterative formulation introduced by Abedin and Kitagawa as the prototype. More importantly, it employs an error tolerance criterion of inexact Aleksandrov solutions to approximately solve the subproblems without spoiling the convergence, which becomes the most crucial issue for the efficient implementation of the iterative method. For the two-dimensional case, by properly taking advantage of the flexibility rendered by the proposed inexact approach and a convergent fixed-point-based approach to solve the subproblems, considerable advancements in computational efficiency can be achieved by the inexact AKI method with its convergence under the ${\cal C}^{2,\alpha}$ boundary condition being rigorously established. Numerical experiments are conducted to demonstrate the efficiency of the proposed inexact AKI method. The numerical results suggest that the inexact AKI method can be more than eight times faster than the original AKI method, at least for all the tested problems.

Aerocapture consists of converting a hyperbolic approach trajectory into a captured target orbit utilizing the aerodynamic forces generated via a single pass through the atmosphere. Aerocapture guidance systems must be robust to significant environmental variations and modeling uncertainty, particularly regarding atmospheric properties and delivery conditions. Recent work has shown that enabling control over both bank angle and angle of attack, a strategy referred to as augmented bank angle modulation (ABAM), can improve robustness to entry state and atmospheric uncertainties. In this work, we derive optimal control solutions for an aerocapture vehicle using ABAM. We first formulate the problem using a linear aerodynamic model and derive closed-form optimal control profiles using Pontryagin's Minimum Principle. To increase modeling fidelity, we also consider a quadratic aerodynamic model and obtain the solution directly using the optimality conditions. Both formulations are solved numerically using Gauss pseudospectral methods (via GPOPS, a software tool for pseudospectral optimal control), to validate the analytic solutions. We then introduce a novel aerocapture guidance algorithm, ABAMGuid+, which indirectly minimizes propellant usage by mimicking the structure of the optimal control solution, enabling efficient guidance by avoiding the complexity of solving the full optimal control problem online. Extensive Monte Carlo simulations of a Uranus aerocapture mission demonstrate that ABAMGuid+ increases capture success rates and reduces post-capture propellant requirements relative to previous methods.

This article analyses the simple projection method proposed by Izuchukwu et al. [8, Algorithm 3.2] for solving variational inequality problems by incorporating momentum terms. A new step size strategy is also introduced, in which the step size sequence increases after a finite number of iterations. Under the assumptions that the underlying operator is quasimonotone and Lipschitz continuous, we establish weak convergence of the proposed method. The effectiveness and efficiency of the algorithm are demonstrated through numerical experiments and are compared with existing approaches from the literature. Finally, we apply the proposed algorithm to a signal recovery problem.

In this paper, the Mean value iterative process is modified with the Mann iterative process for mean nonexpansive mapping in a hyperbolic metric space that satisfy the symmetry criteria and in uniformly convex hyperbolic spaces to validate the iterative process, we present strong and $Delta$-convergence theorems.

In this paper we study the long time behaviour of mean field games systems with fractional diffusion, modeling the case that the individual dynamics of the players is driven by independent jump processes and controlled through the drift term, while being confined by an external field in order to guarantee ergodicity. In the case of globally Lipschitz, locally uniformly convex Hamiltonian, and weakly coupled costs satisfying the Lasry-Lions monotonicity condition, we prove that there is a unique solution $(u_T,m_T)$ to the mean field game problem in $(0,T)$ and we show that, if $T$ is sufficiently large, $(u_T,m_T)$ satisfies the so-called turnpike property, namely it is exponentially close to the (unique) stationary ergodic state for any proportionally long intermediate time.

We consider an evolving random discrete tree model called Preferential Attachment with Vertex Death, as introduced by Deijfen. Initialised with an alive root labelled $1$, at each step $n\geq1$ either a new vertex with label $n+1$ is introduced that attaches to an existing alive vertex selected preferentially according to a function $b$, or an alive vertex is selected preferentially according to a function $d$ and killed. In this article we introduce a generalised concept of persistence for evolving random graph models. Let $O_n$ be the smallest label among all alive vertices (the oldest alive vertex), and let $I_n^m$ be the label of the alive vertex with the $m^{\mathrm{th}}$ largest degree. We say a persistent $m$-hub exists if $I_n^m$ converges almost surely, we say that persistence occurs when $I_n^1/O_n$ is tight, and that lack of persistence occurs when $I_n^1/O_n$ tends to infinity. We identify two regimes called the infinite lifetime and finite lifetime regimes. In the infinite lifetime regime, vertices are never killed with positive probability. Here, we provide conditions under which we prove the (non-)existence of persistent $m$-hubs for any $m\in\mathbb N$. This expands and generalises recent work of Iyer, which covers the case $d\equiv 0$ and $m=1$. In the finite lifetime regime, vertices are killed after a finite number of steps almost surely. Here we provide conditions under which we prove the occurrence of persistence, which complements recent work of Heydenreich and the author, where lack of persistence is studied for preferential attachment with vertex death.

J. Hoste and J. H. Przytycki computed the Kauffman bracket skein module (KBSM) of lens spaces in their papers published in 1993 and 1995. Using a basis for the KBSM of a fibered torus, we construct new bases for the KBSMs of two families of lens spaces: $L(p,2)$ and $L(4k,2k+1)$ with $k\neq 0$. For KBSM of $L(0,1) = {\bf S}^{2}\times S^{1}$, we find a new generating set that yields its decomposition into a direct sum of cyclic modules.

Fourier-transform-based methods enable accurate, dispersion-free simulations of time-domain scattering problems by evaluating solutions to the Helmholtz equation at a discrete set of frequencies, sufficient to approximate the inverse Fourier transform. However, in the case of scattering by trapping obstacles, the Helmholtz solution exhibits nearly-real complex resonances -- which significantly slows the convergence of numerical inverse transform. To address this difficulty this paper introduces a frequency-domain singularity subtraction technique that regularizes the integrand of the inverse transform and efficiently computes the singularity contribution via a combination of a straightforward and inexpensive numerical technique together with a large-time asymptotic expansion. Crucially, all relevant complex resonances and their residues are determined via rational approximation of integral equation solutions at real frequencies. An adaptive algorithm is employed to ensure that all relevant complex resonances are properly identified.

We consider fully discrete numerical approximations for axisymmetric Willmore flow that are unconditionally stable and work reliably without remeshing. We restrict our attention to surfaces without boundary, but allow for spontaneous curvature effects. The axisymmetric setting allows us to formulate our schemes in terms of the generating curve of the considered surface. We propose a novel weak formulation, that combines an evolution equation for the surface's mean curvature and the curvature identity of the generating curve. The mean curvature is used to describe the gradient flow structure, which enables an unconditional stability result for the discrete solutions. The generating curve's curvature, on the other hand, describes the surface's in-plane principal curvature and plays the role of a Lagrange multiplier for an equidistribution property on the discrete level. We introduce two fully discrete schemes and prove their unconditional stability. Numerical results are provided to confirm the convergence, stability and equidistribution properties of the introduced schemes.

We consider computing systems that partition jobs into tasks, add redundancy through coding, and assign the encoded tasks to different computing nodes for parallel execution. The expected execution time depends on the level of redundancy. The computing nodes execute large jobs in batches of tasks. We show that the expected execution time depends on the batch size as well. The optimal batch size that minimizes the execution time depends on the level of redundancy under a fixed number of parallel servers and other system parameters. Furthermore, we show how to (jointly) optimize the redundancy level and batch size to reduce the expected job completion time for two service-time distributions. The simulation presented helps us appreciate the claims.

We derive the spectral domain properties of two-dimensional (2-D) $(\lambda_1, \lambda_2)$-constacyclic codes over $\mathbb{F}_q$ using the 2-D finite field Fourier transform (FFFT). Based on the spectral nulls of 2-D $(\lambda_1, \lambda_2)$-constacyclic codes, we characterize the structure of 2-D constacyclic coded arrays. The proposed 2-D construction has flexible code rates and works for any code areas, be it odd or even area. We present an algorithm to detect the location of 2-D errors. Further, we also propose decoding algorithms for extracting the error values using both time and frequency domain properties by exploiting the sparsity that arises due to duality in the time and frequency domains. Through several illustrative examples, we demonstrate the working of the proposed decoding algorithms.

This work studies optimal control problems of systems with uncertain, probabilistically distributed parameters to optimize average performance. Known as Riemann-Stieltjes, average, or ensemble optimal control, this kind of problem is crucial when parameter uncertainty matters. We derive necessary optimality conditions and characterize feedback controls for control-affine systems. Two scenarios are examined: known initial conditions (finite-dimensional case) and uncertain initial conditions (infinite-dimensional framework). The Pontryagin Maximum Principle is extended using a Hilbert space formulation.

We compute the derivations of Quantum Nilpotent Algebras under a technical (but necessary) assumption on the center. As a consequence, we give an explicit description of the first Hochschild cohomology group of $U_q^+(\mathfrak{g})$, the positive part of the quantized enveloping algebra of a finite-dimensional complex simple Lie algebra $\mathfrak{g}$. Our results are obtained leveraging an initial cluster constructed by Goodearl and Yakimov.

Recently, Clow and McKay proved that the Digraph Placement ruleset is universal for normal play: for all normal play combinatorial games $X$, there is a Digraph Placement game $G$ with $G=X$. Clow and McKay also showed that the 22 game values born by day 2 correspond to Digraph Placement games with at most 4 vertices. This bound is best possible. We extend this work using a combination of exhaustive and random searches to demonstrate all 1474 values born by day 3 correspond to Digraph Placement games on at most 8 vertices. We provide a combinatorial proof that this bound is best possible. We conclude by giving improved bounds on the number of vertices required to construct all game values born by days 4 and 5.

The random current representation has been very successful in analyzing the ferromagnetic Ising model. In this article, we extend the method to the non-ferromagnetic case. To show its usefulness, we derive a covariance identity for the heterogeneous one-dimensional Ising model, which recovers the correct rate of exponential decay of covariances in the case of random i.i.d. interaction and external field.

Security-constrained unit commitment with alternating current optimal power flow (SCUC-ACOPF) is a central problem in power grid operations that optimizes commitment and dispatch of generators under a physically accurate power transmission model while encouraging robustness against component failures. SCUC-ACOPF requires solving large-scale problems that involve multiple time periods and networks with thousands of buses within strict time limits. In this work, we study a detailed SCUC-ACOPF model with a rich set of features of modern power grids, including price-sensitive load, reserve products, transformer controls, and energy-limited devices. We propose a decomposition scheme and a penalty alternating direction method to find high-quality solutions to this model. Our methodology leverages spatial and temporal decomposition, separating the problem into a set of mixed-integer linear programs for each bus and a set of continuous nonlinear programs for each time period. To improve the performance of the algorithm, we introduce a variety of heuristics, including restrictions of temporal linking constraints, a second-order cone relaxation, and a contingency screening algorithm. We quantify the quality of feasible solutions through a dual bound from a convex second-order cone program. To evaluate our algorithm, we use large-scale test cases from Challenge 3 of the U.S. Department of Energy's Grid Optimization Competition that resemble real power grid data under a variety of operating conditions and decision horizons. The experiments yield feasible solutions with an average optimality gap of 1.33%, demonstrating that this approach generates near-optimal solutions within stringent time limits.

We study monogenity of pure cubic number fields by means of Selmer groups of certain elliptic curves. A cubic number field with discriminant $D$ determines a unique nontrivial $\mathbb{F}_3$-orbit in the first cohomology group of the elliptic curve $E^D: y^2 = 4x^3 + D$ with respect to a certain 3-isogeny $\phi$. Orbits corresponding to monogenic fields must lie in the soluble part of the Selmer group $S^{\phi}(E^D/\mathbb{Q})$, and this gives a criterion to discard monogenity. From this, we can derive bounds on the number of monogenic cubic fields in terms of the rank of the elliptic curve. We can also determine the monogenity of many concrete pure cubic fields assuming GRH.

In this paper, we introduce a logarithmic-type second-order model with a non-local logarithmic damping mechanism in $R^N$. We present a motivation with a spectral approach to consider the equation, we consider the Cauchy problem associated with the model. More precisely, we study the asymptotic behavior of solutions as $t$ goes to infinity in $L^2$-sense; namely, we prove results on the asymptotic profile and optimal decay of solutions as time goes to infinity in $L^2$-sense.

We show that for any $\Delta \geq 3$, there is no bound computable from $(\varepsilon, r)$ on the size of a graph required to approximate a graph of maximum degree at most $\Delta$ up to $\varepsilon$ error in $r$-neighborhood statistics. This provides a negative answer to a question posed by Lov\'asz. Our result is a direct consequence of the recent celebrated work of Bowen, Chapman, Lubotzky, and Vidick, which refutes the Aldous--Lyons conjecture.

We study the system of first order PDEs for pseudo-Riemannian metrics governing the Hamiltonian formalism for systems of hydrodynamic type. In the diagonal setting the integrability conditions ensure the compatibility of this system and, thanks to a classical theorem of Darboux, the existence of a family of solutions depending on functional parameters. In this paper we study the generalisation of this result to a class of non-diagonalisable systems of hydrodynamic type that naturally generalises Tsarev's integrable diagonal systems.

In recent years, spiking neural networks (SNNs) have gained momentum due to their high potential in time-series processing combined with minimal energy consumption. However, they still lack a dedicated and efficient training algorithm. The popular backpropagation with surrogate gradients, adapted from stochastic gradient descent (SGD)-derived algorithms, has several drawbacks when used as an optimizer for SNNs. Specifically, it suffers from low scalability and numerical imprecision. In this paper, we propose a novel SNN training method based on the alternating direction method of multipliers (ADMM). Our ADMM-based training aims to solve the problem of the SNN step function's non-differentiability. We formulate the problem, derive closed-form updates, and empirically show the optimizer's convergence properties, great potential, and possible new research directions to improve the method in a simulated proof-of-concept.

For scientific machine learning tasks with a lot of custom code, picking the right Automatic Differentiation (AD) system matters. Our Julia package DifferentiationInterface.jl provides a common frontend to a dozen AD backends, unlocking easy comparison and modular development. In particular, its built-in preparation mechanism leverages the strengths of each backend by amortizing one-time computations. This is key to enabling sophisticated features like sparsity handling without putting additional burdens on the user.

Modular, Jacobi, and mock-modular forms serve as generating functions for BPS black hole degeneracies. By training feed-forward neural networks on Fourier coefficients of automorphic forms derived from the Dedekind eta function, Eisenstein series, and Jacobi theta functions, we demonstrate that machine learning techniques can accurately predict modular weights from truncated expansions. Our results reveal strong performance for negative weight modular and quasi-modular forms, particularly those arising in exact black hole counting formulae, with lower accuracy for positive weights and more complicated combinations of Jacobi theta functions. This study establishes a proof of concept for using machine learning to identify how data is organized in terms of modular symmetries in gravitational systems and suggests a pathway toward automated detection and verification of symmetries in quantum gravity.

We prove the invariance of all planar Feynman graphs 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.

Universal features of chaotic quantum dynamics underlie our understanding of thermalization in closed quantum systems and the complexity of quantum computations. Reversible automaton circuits, comprised of classical logic gates, have emerged as a tractable means to study such dynamics. Despite generating no entanglement in the computational basis, these circuits nevertheless capture many features expected from fully quantum evolutions. In this work, we demonstrate that the differences between automaton dynamics and fully quantum dynamics are revealed by the operator entanglement spectrum, much like the entanglement spectrum of a quantum state distinguishes between the dynamics of states under Clifford and Haar random circuits.While the operator entanglement spectrum under random unitary dynamics is governed by the eigenvalue statistics of random Gaussian matrices, we show evidence that under random automaton dynamics it is described by the statistics of Bernoulli random matrices, whose entries are random variables taking values $0$ or $1$. We study the crossover between automaton and generic unitary operator dynamics as the automaton circuit is doped with gates that introduce superpositions, namely Hadamard or $R_x = e^{-i\frac{\pi}{4}X}$ gates. We find that a constant number of superposition-generating gates is sufficient to drive the operator dynamics to the random-circuit universality class, similar to earlier results on Clifford circuits doped with $T$ gates. This establishes the operator entanglement spectrum as a useful tool for probing the chaoticity and universality class of quantum dynamics.

As sequential learning algorithms are increasingly applied to real life, ensuring data privacy while maintaining their utilities emerges as a timely question. In this context, regret minimisation in stochastic bandits under $\epsilon$-global Differential Privacy (DP) has been widely studied. Unlike bandits without DP, there is a significant gap between the best-known regret lower and upper bound in this setting, though they "match" in order. Thus, we revisit the regret lower and upper bounds of $\epsilon$-global DP algorithms for Bernoulli bandits and improve both. First, we prove a tighter regret lower bound involving a novel information-theoretic quantity characterising the hardness of $\epsilon$-global DP in stochastic bandits. Our lower bound strictly improves on the existing ones across all $\epsilon$ values. Then, we choose two asymptotically optimal bandit algorithms, i.e. DP-KLUCB and DP-IMED, and propose their DP versions using a unified blueprint, i.e., (a) running in arm-dependent phases, and (b) adding Laplace noise to achieve privacy. For Bernoulli bandits, we analyse the regrets of these algorithms and show that their regrets asymptotically match our lower bound up to a constant arbitrary close to 1. This refutes the conjecture that forgetting past rewards is necessary to design optimal bandit algorithms under global DP. At the core of our algorithms lies a new concentration inequality for sums of Bernoulli variables under Laplace mechanism, which is a new DP version of the Chernoff bound. This result is universally useful as the DP literature commonly treats the concentrations of Laplace noise and random variables separately, while we couple them to yield a tighter bound.

This note is an overview of the knot-quiver correspondence, which relates symmetric quivers and their partition functions, a.k.a. motivic Donaldson-Thomas generating series, to quantum invariants of knots and links in $S^3$.

Boundary Discontinuity Designs are used to learn about treatment effects along a continuous boundary that splits units into control and treatment groups according to a bivariate score variable. These research designs are also called Multi-Score Regression Discontinuity Designs, a leading special case being Geographic Regression Discontinuity Designs. We study the statistical properties of commonly used local polynomial treatment effects estimators along the continuous treatment assignment boundary. We consider two distinct approaches: one based explicitly on the bivariate score variable for each unit, and the other based on their univariate distance to the boundary. For each approach, we present pointwise and uniform estimation and inference methods for the treatment effect function over the assignment boundary. Notably, we show that methods based on univariate distance to the boundary exhibit an irreducible large misspecification bias when the assignment boundary has kinks or other irregularities, making the distance-based approach unsuitable for empirical work in those settings. In contrast, methods based on the bivariate score variable do not suffer from that drawback. We illustrate our methods with an empirical application. Companion general-purpose software is provided.

The rapid advancement of LLMs (Large Language Models) has established them as a foundational technology for many AI and ML powered human computer interactions. A critical challenge in this context is the attribution of LLM-generated text, either to the specific language model used or to the individual user who generated it. This is essential for combating misinformation, fake news, misinterpretation, and plagiarism. One of the key techniques for addressing this issue is watermarking. This work presents a watermarking scheme for LLM-generated text based on Lagrange interpolation, which enables the recovery of a secret author identity even when the text has been heavily redacted by an adversary. The core idea is to embed a continuous sequence of points (x, f(x)) that lie on a single straight line. The x-coordinates are generated pseudorandomly using either an LFSR (when security is not a priority) or a cryptographically secure NFSR for high-security applications. The scheme efficiency and resilience to adversarial modifications are analysed. Experimental results show that the proposed method is highly effective, allowing the recovery of the author identity when as few as three points survive adversarial manipulation.

Urban traffic congestion, exacerbated by inefficient parking management and cruising for parking, significantly hampers mobility and sustainability in smart cities. Drivers often face delays searching for parking spaces, influenced by factors such as accessibility, cost, distance, and available services such as charging facilities in the case of electric vehicles. These inefficiencies contribute to increased urban congestion, fuel consumption, and environmental impact. Addressing these challenges, this paper proposes a feedback control incentivisation-based system that aims to better distribute vehicles between city and suburban parking facilities offering park-and-charge/-ride services. Individual driver behaviours are captured via discrete choice models incorporating factors of importance to parking location choice among drivers, such as distance to work, public transport connectivity, charging infrastructure availability, and amount of incentive offered; and are regulated through principles of ergodic control theory. The proposed framework is applied to an electric vehicle park-and-charge/-ride problem, and demonstrates how predictable long-term behaviour of the system can be guaranteed.

Heating, Ventilation, and Air Conditioning (HVAC) systems account for approximately 38% of building energy consumption globally, making them one of the most energy-intensive services. The increasing emphasis on energy efficiency and sustainability, combined with the need for enhanced occupant comfort, presents a significant challenge for traditional HVAC systems. These systems often fail to dynamically adjust to real-time changes in electricity market rates or individual comfort preferences, leading to increased energy costs and reduced comfort. In response, we propose a Human-in-the-Loop (HITL) Artificial Intelligence framework that optimizes HVAC performance by incorporating real-time user feedback and responding to fluctuating electricity prices. Unlike conventional systems that require predefined information about occupancy or comfort levels, our approach learns and adapts based on ongoing user input. By integrating the occupancy prediction model with reinforcement learning, the system improves operational efficiency and reduces energy costs in line with electricity market dynamics, thereby contributing to demand response initiatives. Through simulations, we demonstrate that our method achieves significant cost reductions compared to baseline approaches while maintaining or enhancing occupant comfort. This feedback-driven approach ensures personalized comfort control without the need for predefined settings, offering a scalable solution that balances individual preferences with economic and environmental goals.

We investigate privacy-preserving spectral clustering for community detection within stochastic block models (SBMs). Specifically, we focus on edge differential privacy (DP) and propose private algorithms for community recovery. Our work explores the fundamental trade-offs between the privacy budget and the accurate recovery of community labels. Furthermore, we establish information-theoretic conditions that guarantee the accuracy of our methods, providing theoretical assurances for successful community recovery under edge DP.

In the last few decades, Machine Learning (ML) has achieved significant success across domains ranging from healthcare, sustainability, and the social sciences, to criminal justice and finance. But its deployment in increasingly sophisticated, critical, and sensitive areas affecting individuals, the groups they belong to, and society as a whole raises critical concerns around fairness, transparency, robustness, and privacy, among others. As the complexity and scale of ML systems and of the settings in which they are deployed grow, so does the need for responsible ML methods that address these challenges while providing guaranteed performance in deployment. Mixed-integer optimization (MIO) offers a powerful framework for embedding responsible ML considerations directly into the learning process while maintaining performance. For example, it enables learning of inherently transparent models that can conveniently incorporate fairness or other domain specific constraints. This tutorial paper provides an accessible and comprehensive introduction to this topic discussing both theoretical and practical aspects. It outlines some of the core principles of responsible ML, their importance in applications, and the practical utility of MIO for building ML models that align with these principles. Through examples and mathematical formulations, it illustrates practical strategies and available tools for efficiently solving MIO problems for responsible ML. It concludes with a discussion on current limitations and open research questions, providing suggestions for future work.

A generalization of the notion of spanning tree congestion for weighted graphs is introduced. The $L^p$ congestion of a spanning tree is defined as the $L^p$ norm of the edge congestion of that tree. In this context, the classical congestion is the $L^\infty$-congestion. Explicit estimations of the minimal spanning tree $L^p$ congestion for some families of graphs are given. In addition, we introduce a polynomial-time algorithm for approximating the minimal $L^p$-congestion spanning tree in any weighted graph and another two similar algorithms for weighted planar graphs. The performance of these algorithms is tested in several graphs.

In this letter, we investigate the transmission of a complex-valued parameter vector from a transmitter to an intended receiver, considering both the presence and absence of an eavesdropper. The direct links from the transmitter to both the intended receiver and the eavesdropper are assumed to be blocked, and communications occur solely through cascaded channels facilitated by a beyond-diagonal reconfigurable intelligent surface (BD-RIS). While previous research has considered this system under conventional (diagonal) RIS assistance, we extend the setup to incorporate BD-RIS and quantify the resulting improvement in estimation performance at the intended receiver. This performance is measured by the trace of the Fisher information matrix (FIM), or equivalently, the average Fisher information, while simultaneously limiting the estimation capability of the eavesdropper. We propose solutions and algorithms for optimizing the BD-RIS response matrix and demonstrate their effectiveness. Numerical results reveal that the BD-RIS provides a significant enhancement in estimation quality compared to conventional diagonal RIS architectures.

We present a quantum variational algorithm based on a novel circuit that generates all permutations that can be spanned by one- and two-qubits permutation gates. The construction of the circuits follows from group-theoretical results, most importantly the Bruhat decomposition of the group generated by the \(\mathtt{cx}\) gates. These circuits require a number of qubits that scale logarithmically with the permutation dimension, and are therefore employable in near-term applications. We further augment the circuits with ancilla qubits to enlarge their span, and with these we build ansatze to tackle permutation-based optimization problems such as quadratic assignment problems, and graph isomorphisms. The resulting quantum algorithm, \textsc{QuPer}, is competitive with respect to classical heuristics and we could simulate its behavior up to a problem with $256$ variables, requiring $20$ qubits.

Robust self-testing in non-local games allows a classical referee to certify that two untrustworthy players are able to perform a specific quantum strategy up to high precision. Proving robust self-testing results becomes significantly easier when one restricts the allowed strategies to symmetric projective maximally entangled (PME) strategies, which allow natural descriptions in terms of tracial von Neumann algebras. This has been exploited in the celebrated MIP*=RE paper and related articles to prove robust self-testing results for synchronous games when restricting to PME strategies. However, the PME assumptions are not physical, so these results need to be upgraded to make them physically relevant. In this work, we do just that: we prove that any perfect synchronous game which is a robust self-test when restricted to PME strategies, is in fact a robust self-test for all strategies. We then apply our result to the Quantum Low Degree Test to find an efficient $n$-qubit test.

As shown by Robertson and Seymour, deciding whether the complete graph $K_t$ is a minor of an input graph $G$ is a fixed parameter tractable problem when parameterized by $t$. From the approximation viewpoint, the gap to fill is quite large, as there is no PTAS for finding the largest complete minor unless $P = NP$, whereas a polytime $O(\sqrt n)$-approximation algorithm was given by Alon, Lingas and Wahl\'en. We investigate the complexity of finding $K_t$ as interval minor in ordered graphs (i.e. graphs with a linear order on the vertices, in which intervals are contracted to form minors). Our main result is a polytime $f(t)$-approximation algorithm, where $f$ is triply exponential in $t$ but independent of $n$. The algorithm is based on delayed decompositions and shows that ordered graphs without a $K_t$ interval minor can be constructed via a bounded number of three operations: closure under substitutions, edge union, and concatenation of a stable set. As a byproduct, graphs avoiding $K_t$ as an interval minor have bounded chromatic number.

The approximation capabilities of Deep Q-Networks (DQNs) are commonly justified by general Universal Approximation Theorems (UATs) that do not leverage the intrinsic structural properties of the optimal Q-function, the solution to a Bellman equation. This paper establishes a UAT for a class of DQNs whose architecture is designed to emulate the iterative refinement process inherent in Bellman updates. A central element of our analysis is the propagation of regularity: while the transformation induced by a single Bellman operator application exhibits regularity, for which Backward Stochastic Differential Equations (BSDEs) theory provides analytical tools, the uniform regularity of the entire sequence of value iteration iterates--specifically, their uniform Lipschitz continuity on compact domains under standard Lipschitz assumptions on the problem data--is derived from finite-horizon dynamic programming principles. We demonstrate that layers of a deep residual network, conceived as neural operators acting on function spaces, can approximate the action of the Bellman operator. The resulting approximation theorem is thus intrinsically linked to the control problem's structure, offering a proof technique wherein network depth directly corresponds to iterations of value function refinement, accompanied by controlled error propagation. This perspective reveals a dynamic systems view of the network's operation on a space of value functions.

We study the probability that a given candidate is an alpha-winner, i.e. a candidate preferred to each other candidate j by a fraction alpha_j of the voters. This extends the classical notion of Condorcet winner, which corresponds to the case alpha = (1/2, ..., 1/2). Our analysis is conducted under the general assumption that voters have independent preferences, illustrated through applications to well-known models such as Impartial Culture and the Mallows model. While previous works use probabilistic arguments to derive the limiting probability as the number of voters tends to infinity, we employ techniques from the field of analytic combinatorics to compute convergence rates and provide a method for obtaining higher-order terms in the asymptotic expansion. In particular, we establish that the probability of a given candidate being the Condorcet winner in Impartial Culture is a_0 + a_{1, n} n^{-1/2} + O(n^{-1}), where we explicitly provide the values of the constant a_0 and the coefficient a_{1, n}, which depends solely on the parity of the number of voters n. Along the way, we derive technical results in multivariate analytic combinatorics that may be of independent interest.

An ant-like observer confined to a two-dimensional surface traversed by stripes would wonder whether this striped landscape could be devised in such a way as to appear to be the same wherever they go. Differently stated, this is the problem studied in this paper. In a more technical jargon, we determine all possible uniform nematic fields on a smooth surface. It was already known that for such a field to exist, the surface must have constant negative Gaussian curvature. Here, we show that all uniform nematic fields on such a surface are parallel transported (in Levi-Civita's sense) by special systems of geodesics, which (with scant inventiveness) are termed uniform. We prove that, for every geodesic on the surface, there are two systems of uniform geodesics that include it; they are conventionally called right and left, to allude at a possible intrinsic definition of handedness. We found explicitly all uniform fields for Beltrami's pseudosphere. Since both geodesics and uniformity are preserved under isometries, by a classical theorem of Minding, the solution for the pseudosphere carries over all other admissible surfaces, thus providing a general solution to the problem (at least in principle).

Federated learning faces severe communication bottlenecks due to the high dimensionality of model updates. Communication compression with contractive compressors (e.g., Top-K) is often preferable in practice but can degrade performance without proper handling. Error feedback (EF) mitigates such issues but has been largely restricted for smooth, unconstrained problems, limiting its real-world applicability where non-smooth objectives and safety constraints are critical. We advance our understanding of EF in the canonical non-smooth convex setting by establishing new lower complexity bounds for first-order algorithms with contractive compression. Next, we propose Safe-EF, a novel algorithm that matches our lower bound (up to a constant) while enforcing safety constraints essential for practical applications. Extending our approach to the stochastic setting, we bridge the gap between theory and practical implementation. Extensive experiments in a reinforcement learning setup, simulating distributed humanoid robot training, validate the effectiveness of Safe-EF in ensuring safety and reducing communication complexity.

This paper addresses the efficient computation of Jacobian matrices for programs composed of sequential differentiable subprograms. By representing the overall Jacobian as a chain product of the Jacobians of these subprograms, we reduce the problem to optimizing the sequence of matrix multiplications, known as the Jacobian Matrix Chain Product problem. Solutions to this problem yield "optimal bracketings", which induce a precedence-constraint scheduling problem. We investigate the inherent parallelism in the solutions and develop a new dynamic programming algorithm as a heuristic that incorporates the scheduling. To assess its performance, we benchmark it against the global optimum, which is computed via a branch-and-bound algorithm.

Hydrodynamic dispersion determines the extent of solute mixing and apparent diffusive flux in various chemical, geological, and biological systems that possess flow with strong velocity gradients. Though established dispersion theory captures the transient interplay between diffusion, flow, and reaction rates, a long-time approximation is commonly adopted where the characteristic time scale of flow substantially exceeds that of pore-scale diffusion. Here, we introduce a frequency-based theory to model omni-temporal dispersion through porous media that simultaneously captures fast and slow components of dispersion. To create this theory, we formally volume-average the Fourier-transformed pore-scale advection-diffusion equation for solute and obtain up-scaled transport coefficients for a periodic unit cell: a dispersion tensor, an advection suppression vector, and reaction-rate coefficients. These coefficients serve dually as transfer functions that correlate certain output with input quantities in the frequency domain, enabling the prediction of up-scaled temporal dynamics using inverse Fourier transformation. Their utility is demonstrated by deriving analytical expressions for the dispersion coefficient for cases of Poiseuille flow between parallel plates and through circular tubes. This omni-temporal theory produces breakthrough curves for a fast solute pulse between inactive parallel plates that agree with the solute propagation rates predicted by direct numerical simulation (DNS), in contrast with conventional asymptotic theory that overpredicts such by orders of magnitude. Deviations in solute peak magnitude from DNS are shown to arise from solute back-diffusion into the inlet plane as well as entry-region effects that are evidenced by a non-periodic variation of the closure variable b.

Quantum error correction (QEC) is crucial for numerous quantum applications, including fault-tolerant quantum computation, which is of great scientific and industrial interest. Among various QEC paradigms, topological quantum error correction (TQEC) has attained the most experimental successes by far. In this paper, we build upon existing knowledge of TQEC by developing a generalized theoretical framework of TQEC. We begin by formally defining TQEC codes and exploring the algebraic topological principles underlying these quantum codes, including deriving the conditions for any topological manifold to serve as a quantum memory. We show that TQEC for qubits works for both orientable and non-orientable manifolds. Moreover, we extend the construction of TQEC to higher-dimensional manifolds and provide examples for higher-dimensional TQEC codes. Finally, we apply these principles to construct new codes on 2-dimensional manifolds that have received limited attention in prior literature. As a case study, we simulate the performance of TQEC codes on the Klein bottle $K$ and evaluate their efficacy for quantum error correction. This work contributes to the advancement of TQEC by proposing a broader class of codes and demonstrating their theoretical and practical potential. By addressing previously unexplored topological structures, our findings represent a step forward in achieving fault-tolerant quantum computation and communication.

At present, there is an abundance of results showing that the phase-field approach to fracture in elastic brittle materials -- when properly accounting for material strength -- describes the \emph{nucleation} of fracture from large pre-existing cracks in a manner that is consistent with the Griffith competition between bulk deformation energy and surface fracture energy. By contrast, results that demonstrate the ability of this approach to describe Griffith fracture \emph{propagation} are scarce and restricted to Mode I. Aimed at addressing this lacuna, the main objective of this paper is to show that the phase-field approach to fracture describes Mode III fracture propagation in a manner that is indeed consistent with the Griffith energy competition. This is accomplished via direct comparisons between phase-field predictions for fracture propagation in the so-called \emph{trousers} \emph{test} and the corresponding results that emerge from the Griffith energy competition. The latter are generated from full-field finite-element solutions that -- as an additional critical contribution of this paper -- also serve to bring to light the hitherto unexplored limitations of the classical Rivlin-Thomas-Greensmith formulas that are routinely used to analyze the trousers test.

Generalized bicycle (GB) codes have emerged as a promising class of quantum error-correcting codes with practical decoding capabilities. While numerous asymptotically good quantum codes and quantum low-density parity-check code constructions have been proposed, their finite block-length performance often remains unquantified. In this work, we demonstrate that GB codes exhibit comparable or superior error correction performance in finite-length settings, particularly when designed with higher or unrestricted row weights. Leveraging their flexible construction, GB codes can be tailored to achieve high rates while maintaining efficient decoding. We evaluate GB codes against other leading quantum code families, such as quantum Tanner codes and single-parity-check product codes, highlighting their versatility in practical finite-length applications.

We study two models of ReLU neural networks: monotone networks (ReLU$^+$) and input convex neural networks (ICNN). Our focus is on expressivity, mostly in terms of depth, and we prove the following lower bounds. For the maximum function MAX$_n$ computing the maximum of $n$ real numbers, we show that ReLU$^+$ networks cannot compute MAX$_n$, or even approximate it. We prove a sharp $n$ lower bound on the ICNN depth complexity of MAX$_n$. We also prove depth separations between ReLU networks and ICNNs; for every $k$, there is a depth-2 ReLU network of size $O(k^2)$ that cannot be simulated by a depth-$k$ ICNN. The proofs are based on deep connections between neural networks and polyhedral geometry, and also use isoperimetric properties of triangulations.

Integrated autoregressive conditional duration (ACD) models serve as natural counterparts to the well-known integrated GARCH models used for financial returns. However, despite their resemblance, asymptotic theory for ACD is challenging and also not complete, in particular for integrated ACD. Central challenges arise from the facts that (i) integrated ACD processes imply durations with infinite expectation, and (ii) even in the non-integrated case, conventional asymptotic approaches break down due to the randomness in the number of durations within a fixed observation period. Addressing these challenges, we provide here unified asymptotic theory for the (quasi-) maximum likelihood estimator for ACD models; a unified theory which includes integrated ACD models. Based on the new results, we also provide a novel framework for hypothesis testing in duration models, enabling inference on a key empirical question: whether durations possess a finite or infinite expectation. We apply our results to high-frequency cryptocurrency ETF trading data. Motivated by parameter estimates near the integrated ACD boundary, we assess whether durations between trades in these markets have finite expectation, an assumption often made implicitly in the literature on point process models. Our empirical findings indicate infinite-mean durations for all the five cryptocurrencies examined, with the integrated ACD hypothesis rejected -- against alternatives with tail index less than one -- for four out of the five cryptocurrencies considered.