Detection of pulmonary nodules by CT is used for screening lung cancer in early stages.omputer aided diagnosis (CAD) based on deep-learning method can identify the suspected areas of pulmonary nodules in CT images, thus improving the accuracy and efficiency of CT diagnosis. The accuracy and robustness of deep learning models. Method:In this paper, we explore (1) the data augmentation method based on the generation model and (2) the model structure improvement method based on the embedding mechanism. Two strategies have been introduced in this study: a new data augmentation method and a embedding mechanism. In the augmentation method, a 3D pixel-level statistics algorithm is proposed to generate pulmonary nodule and by combing the faked pulmonary nodule and healthy lung, we generate new pulmonary nodule samples. The embedding mechanism are designed to better understand the meaning of pixels of the pulmonary nodule samples by introducing hidden variables. Result: The result of the 3DVNET model with the augmentation method for pulmonary nodule detection shows that the proposed data augmentation method outperforms the method based on generative adversarial network (GAN) framework, training accuracy improved by 1.5%, and with embedding mechanism for pulmonary nodules classification shows that the embedding mechanism improves the accuracy and robustness for the classification of pulmonary nodules obviously, the model training accuracy is close to 1 and the model testing F1-score is 0.90.Conclusion:he proposed data augmentation method and embedding mechanism are beneficial to improve the accuracy and robustness of the model, and can be further applied in other common diagnostic imaging tasks.

Multiple exposures, of a single illuminated non-configurable mask that is transversely displaced to a number of specified positions, can be used to create any desired distribution of radiant exposure. An experimental proof-of-concept is given for this idea, employing hard X rays. The method is termed "ghost projection", since it may be viewed as a reversed form of classical ghost imaging. The written pattern is arbitrary, up to a tunable constant offset, together with a limiting spatial resolution that is governed by the finest features present in the illuminated mask. The method, which is immune to both proximity-correction and aspect-ratio issues, can be used to make a universal lithographic mask in the hard-X-ray regime. Ghost projection may also be used as a dynamically-configurable beam-shaping element, namely the hard-X-ray equivalent of a spatial light modulator. The idea may be applied to other forms of radiation and matter waves, such as gamma rays, neutrons, electrons, muons, and atomic beams.

This paper studies the leader-following consensuses of uncertain and nonlinear multi-agent systems against composite attacks (CAs), including Denial of Service (DoS) attacks and actuation attacks (AAs). A double-layer control framework is formulated, where a digital twin layer (TL) is added beside the traditional cyber-physical layer (CPL), inspired by the recent Digital Twin technology. Consequently, the resilient control task against CAs can be divided into two parts: One is distributed estimation against DoS attacks on the TL and the other is resilient decentralized tracking control against actuation attacks on the CPL. %The data-driven scheme is used to deal with both model non-linearity and model uncertainty, in which only the input and output data of the system are employed throughout the whole control process. First, a distributed observer based on switching estimation law against DoS is designed on TL. Second, a distributed model free adaptive control (DMFAC) protocol based on attack compensation against AAs is designed on CPL. Moreover, the uniformly ultimately bounded convergence of consensus error of the proposed double-layer DMFAC algorithm is strictly proved. Finally, the simulation verifies the effectiveness of the resilient double-layer control scheme.

Deep learning (DL) has been extensively researched in the field of computed tomography (CT) reconstruction with incomplete data, particularly in sparse-view CT reconstruction. However, applying DL to sparse-view cone beam CT (CBCT) remains challenging. Many models learn the mapping from sparse-view CT images to ground truth but struggle to achieve satisfactory performance in terms of global artifact removal. Incorporating sinogram data and utilizing dual-domain information can enhance anti-artifact performance, but this requires storing the entire sinogram in memory. This presents a memory issue for high-resolution CBCT sinograms, limiting further research and application. In this paper, we propose a cube-based 3D denoising diffusion probabilistic model (DDPM) for CBCT reconstruction using down-sampled data. A DDPM network, trained on cubes extracted from paired fully sampled sinograms and down-sampled sinograms, is employed to inpaint down-sampled sinograms. Our method divides the entire sinogram into overlapping cubes and processes these cubes in parallel using multiple GPUs, overcoming memory limitations. Experimental results demonstrate that our approach effectively suppresses few-view artifacts while preserving textural details faithfully.

After the loss of control authority over thrusters of the Nauka module, the International Space Station lost attitude control for 45 minutes with potentially disastrous consequences. Motivated by a scenario of orbital inspection, we consider a similar malfunction occurring to the inspector satellite and investigate whether its mission can still be safely fulfilled. While a natural approach is to counteract in real-time the uncontrolled and undesirable thrust with the remaining controlled thrusters, vehicles are often subject to actuation delays hindering this approach. Instead, we extend resilience theory to systems suffering from actuation delay and build a resilient trajectory tracking controller with stability guarantees relying on a state predictor. We demonstrate that this controller can track accurately the reference trajectory of the inspection mission despite the actuation delay and the loss of control authority over one of the thrusters.

The integrated use of non-terrestrial network (NTN) entities such as the high-altitude platform station (HAPS) and low-altitude platform station (LAPS) has become essential elements in the space-air-ground integrated networks (SAGINs). However, the complexity, mobility, and heterogeneity of NTN entities and resources present various challenges from system design to deployment. This paper proposes a novel approach to designing a heterogeneous network consisting of HAPSs and unmanned aerial vehicles (UAVs) being LAPS entities. Our approach involves jointly optimizing the three-dimensional trajectory and channel allocation for aerial base stations, with a focus on ensuring fairness and the provision of quality of service (QoS) to ground users. Furthermore, we consider the load on base stations and incorporate this information into the optimization problem. The proposed approach utilizes a combination of deep reinforcement learning and fixed-point iteration techniques to determine the UAV locations and channel allocation strategies. Simulation results reveal that our proposed deep learning-based approach significantly outperforms learning-based and conventional benchmark models.

The deployment of low earth orbit (LEO) satellites with terrestrial networks can potentially increase the efficiency and reduce the cost of relaying content from a data center to a set of edge caches hosted by 6G and beyond enabled macro base stations. In this work, the characteristics of the communication system and the mobility of LEO satellites are thoroughly discussed to describe the channel characteristics of LEO satellites, in terms of their frequency bands, latency, Doppler shifts, fading effects, and satellite access. Three different scenarios are proposed for the relay of data from data centers to edge caches via LEO satellites, which are the "Immediate Forward", "Relay and Forward", and "Store and Forward" scenarios. A comparative problem formulation is utilized to obtain numerical results from simulations to demonstrate the effectiveness and validity as well as the trade-offs of the proposed system model. The simulation results indicate that the integration of LEO satellites in edge caching for 6G and beyond networks decreased the required transmission power for relaying the data from the data center to the edge caches. Future research directions based on the proposed model are discussed.

We show that training a multi-headed self-attention-based deep network to predict deleted, information-dense 2-8 Hz speech modulations over a 1.5-second section of a speech utterance is an effective way to make machines learn to extract speech modulations using time-domain contextual information. Our work exhibits that, once trained on large volumes of unlabelled data, the outputs of the self-attention layers vary in time with a modulation peak at 4 Hz. These pre-trained layers can be used to initialize parts of an Automatic Speech Recognition system to reduce its reliance on labeled speech data greatly.

This work extends the multiscale structure originally developed for point cloud geometry compression to point cloud attribute compression. To losslessly encode the attribute while maintaining a low bitrate, accurate probability prediction is critical. With this aim, we extensively exploit cross-scale, cross-group, and cross-color correlations of point cloud attribute to ensure accurate probability estimation and thus high coding efficiency. Specifically, we first generate multiscale attribute tensors through average pooling, by which, for any two consecutive scales, the decoded lower-scale attribute can be used to estimate the attribute probability in the current scale in one shot. Additionally, in each scale, we perform the probability estimation group-wisely following a predefined grouping pattern. In this way, both cross-scale and (same-scale) cross-group correlations are exploited jointly. Furthermore, cross-color redundancy is removed by allowing inter-color processing for YCoCg/RGB alike multi-channel attributes. The proposed method not only demonstrates state-of-the-art compression efficiency with significant performance gains over the latest G-PCC on various contents but also sustains low complexity with affordable encoding and decoding runtime.

Most approaches for self-triggered control (STC) of nonlinear networked control systems (NCS) require measurements of the full system state to determine transmission times. However, for most control systems only a lower dimensional output is available. To bridge this gap, we present in this paper an output-feedback STC approach for nonlinear NCS. An asymptotically stable observer is used to reconstruct the plant state and transmission times are determined based on the observer state. The approach employs hybrid Lyapunov functions and a dynamic variable to encode past state information and to maximize the time between transmissions. It is non-conservative in the sense that the assumptions on plant and controller are the same as for dynamic STC based on hybrid Lyapunov functions with full state measurements and any asymptotically stabilizing observer can be used. We conclude that the proposed STC approach guarantees asymptotic stability of the origin for the closed-loop system.

Two sound field reproduction methods, weighted pressure matching and weighted mode matching, are theoretically and experimentally compared. The weighted pressure and mode matching are a generalization of conventional pressure and mode matching, respectively. Both methods are derived by introducing a weighting matrix in the pressure and mode matching. The weighting matrix in the weighted pressure matching is defined on the basis of the kernel interpolation of the sound field from pressure at a discrete set of control points. In the weighted mode matching, the weighting matrix is defined by a regional integration of spherical wavefunctions. It is theoretically shown that the weighted pressure matching is a special case of the weighted mode matching by infinite-dimensional harmonic analysis for estimating expansion coefficients from pressure observations. The difference between the two methods are discussed through experiments.

Deep learning models have shown promising performance in the field of diabetic retinopathy (DR) staging. However, collaboratively training a DR staging model across multiple institutions remains a challenge due to non-iid data, client reliability, and confidence evaluation of the prediction. To address these issues, we propose a novel federated uncertainty-aware aggregation paradigm (FedUAA), which considers the reliability of each client and produces a confidence estimation for the DR staging. In our FedUAA, an aggregated encoder is shared by all clients for learning a global representation of fundus images, while a novel temperature-warmed uncertainty head (TWEU) is utilized for each client for local personalized staging criteria. Our TWEU employs an evidential deep layer to produce the uncertainty score with the DR staging results for client reliability evaluation. Furthermore, we developed a novel uncertainty-aware weighting module (UAW) to dynamically adjust the weights of model aggregation based on the uncertainty score distribution of each client. In our experiments, we collect five publicly available datasets from different institutions to conduct a dataset for federated DR staging to satisfy the real non-iid condition. The experimental results demonstrate that our FedUAA achieves better DR staging performance with higher reliability compared to other federated learning methods. Our proposed FedUAA paradigm effectively addresses the challenges of collaboratively training DR staging models across multiple institutions, and provides a robust and reliable solution for the deployment of DR diagnosis models in real-world clinical scenarios.

We propose an open loop methodology based on sample statistics to solve chance constrained stochastic optimal control problems with probabilistic safety guarantees for linear systems where the additive Gaussian noise has unknown mean and covariance. We consider a joint chance constraint for time-varying polytopic target sets under assumptions that the disturbance has been sufficiently sampled. We derive two theorems that allow us to bound the probability of the state being more than some number of sample standard deviations away from the sample mean. We use these theorems to reformulate the chance constraint into a series of convex and linear constraints. Here, solutions guarantee chance constraint satisfaction. We demonstrate our method on a satellite rendezvous maneuver and provide comparisons with the scenario approach.

The scope of this study is to develop a novel sensorless adaptive vibration suppression controller for two-mass systems with joint estimation of states and controller parameters. Unlike existing solutions, we simultaneously: (i) propose an analytically proved, unified and singularity-issue-free scheme of parameters adjustment of a control law with additional feedbacks that ensures convergence of such parameters to their true values under extremely weak regressor finite excitation (FE) requirement, (ii) derive an adaptive observer of a two-mass electromechanical system physical states with guarantee of their convergence to the ground truth values under clear FE condition, (iii) rigorously prove the exponential stability of the obtained closed-loop system of adaptive vibration suppression for two-mass systems that includes the above-mentioned adaptive observer and adaptive controller. These approaches are grounded on the recently proposed method of parameters identification for one class of nonlinearly parameterized regression equation and thoroughly investigated dynamic regression extension and mixing procedure (DREM). The obtained theoretical results are confirmed via numerical experiments.

Blind image quality assessment (BIQA) is a task that predicts the perceptual quality of an image without its reference. Research on BIQA attracts growing attention due to the increasing amount of user-generated images and emerging mobile applications where reference images are unavailable. The problem is challenging due to the wide range of content and mixed distortion types. Many existing BIQA methods use deep neural networks (DNNs) to achieve high performance. However, their large model sizes hinder their applicability to edge or mobile devices. To meet the need, a novel BIQA method with a small model, low computational complexity, and high performance is proposed and named "GreenBIQA" in this work. GreenBIQA includes five steps: 1) image cropping, 2) unsupervised representation generation, 3) supervised feature selection, 4) distortion-specific prediction, and 5) regression and decision ensemble. Experimental results show that the performance of GreenBIQA is comparable with that of state-of-the-art deep-learning (DL) solutions while demanding a much smaller model size and significantly lower computational complexity.

A sample identifying complexity and a sample deciphering time have been introduced in a previous study to capture an estimation error and a computation time of system identification by adversaries. The quantities play a crucial role in defining the security of encrypted control systems and designing a security parameter. This study proposes an optimal security parameter for an encrypted control system under a network eavesdropper and a malicious controller server who attempt to identify system parameters using a least squares method. The security parameter design is achieved based on a modification of conventional homomorphic encryption for improving a sample deciphering time and a novel sample identifying complexity, characterized by controllability Gramians and the variance ratio of identification input to system noise. The effectiveness of the proposed design method for a security parameter is demonstrated through numerical simulations.

In this study, a small, inexpensive remotely driven underwater vehicle that can navigate in shallow water for the purpose of monitoring water quality and demonstrating vehicle control algorithms is presented. The vehicle is operated by an onboard micro-controller, and the sensor payload comprises a turbidity sensor for determining the quality of the water, a depth sensor, and a 9-axis inertial measurement unit. The developed vehicle is an open frame remotely operated vehicle (ROV) with a small footprint and a modular physical and electrical architecture. With a net weight of 1.6 kg, a maximum depth rating of 20 meters, and a development cost of around $80, the ROV frame is composed of polyvinyl chloride tubes and has a length of 0.35 meters. As a ground station, a dedicated laptop shows crucial vehicle data in real time and can send commands to the vehicle. Initial testing in the pool demonstrates that the vehicle is completely operational and effectively complies with pilot commands.

In this article, a simplified modeling and system identification procedure for yaw motion of an unmanned surface vehicle (USV) is presented. Two thrusters that allow for both speed and direction control propel the USV. The outputs of the vehicle under inquiry include parameters that define the mobility of the USV in horizontal plane, such as yaw angle and yaw rate. A linear second order model is first developed, and the unknown coefficients are then determined using data from pool trials. Finally, simulations are carried out to verify the model so that it may be used in a later study to implement various control algorithms.

This paper considers the localization problem in a 5G-aided global navigation satellite system (GNSS) based on real-time kinematic (RTK) technique. Specifically, the user's position is estimated based on the hybrid measurements, including GNSS pseudo-ranges, GNSS carrier phases, 5G angle-of-departures, and 5G channel delays. The underlying estimation problem is solved by steps that comprise obtaining the float solution, ambiguity resolution, and resolving the fixed solution. The analysis results show that the involvement of 5G observations can enable localization under satellite-deprived environments, inclusive of extreme cases with only 2 or 3 visible satellites. Moreover, extensive simulation results reveal that with the help of 5G observations, the proposed algorithm can significantly reduce the estimation error of the user's position and increase the success rate of carrier-phase ambiguity resolution.

Cell detection is a fundamental task in computational pathology that can be used for extracting high-level medical information from whole-slide images. For accurate cell detection, pathologists often zoom out to understand the tissue-level structures and zoom in to classify cells based on their morphology and the surrounding context. However, there is a lack of efforts to reflect such behaviors by pathologists in the cell detection models, mainly due to the lack of datasets containing both cell and tissue annotations with overlapping regions. To overcome this limitation, we propose and publicly release OCELOT, a dataset purposely dedicated to the study of cell-tissue relationships for cell detection in histopathology. OCELOT provides overlapping cell and tissue annotations on images acquired from multiple organs. Within this setting, we also propose multi-task learning approaches that benefit from learning both cell and tissue tasks simultaneously. When compared against a model trained only for the cell detection task, our proposed approaches improve cell detection performance on 3 datasets: proposed OCELOT, public TIGER, and internal CARP datasets. On the OCELOT test set in particular, we show up to 6.79 improvement in F1-score. We believe the contributions of this paper, including the release of the OCELOT dataset at https://lunit-io.github.io/research/publications/ocelot are a crucial starting point toward the important research direction of incorporating cell-tissue relationships in computation pathology.

The advent of Vision Transformer (ViT) has brought substantial advancements in 3D volumetric benchmarks, particularly in 3D medical image segmentation. Concurrently, Multi-Layer Perceptron (MLP) networks have regained popularity among researchers due to their comparable results to ViT, albeit with the exclusion of the heavy self-attention module. This paper introduces a permutable hybrid network for volumetric medical image segmentation, named PHNet, which exploits the advantages of convolution neural network (CNN) and MLP. PHNet addresses the intrinsic isotropy problem of 3D volumetric data by utilizing both 2D and 3D CNN to extract local information. Besides, we propose an efficient Multi-Layer Permute Perceptron module, named MLPP, which enhances the original MLP by obtaining long-range dependence while retaining positional information. Extensive experimental results validate that PHNet outperforms the state-of-the-art methods on two public datasets, namely, COVID-19-20 and Synapse. Moreover, the ablation study demonstrates the effectiveness of PHNet in harnessing the strengths of both CNN and MLP. The code will be accessible to the public upon acceptance.

Knee OsteoArthritis (KOA) is a prevalent musculoskeletal disorder that causes decreased mobility in seniors. The diagnosis provided by physicians is subjective, however, as it relies on personal experience and the semi-quantitative Kellgren-Lawrence (KL) scoring system. KOA has been successfully diagnosed by Computer-Aided Diagnostic (CAD) systems that use deep learning techniques like Convolutional Neural Networks (CNN). In this paper, we propose a novel Siamese-based network, and we introduce a new hybrid loss strategy for the early detection of KOA. The model extends the classical Siamese network by integrating a collection of Global Average Pooling (GAP) layers for feature extraction at each level. Then, to improve the classification performance, a novel training strategy that partitions each training batch into low-, medium- and high-confidence subsets, and a specific hybrid loss function are used for each new label attributed to each sample. The final loss function is then derived by combining the latter loss functions with optimized weights. Our test results demonstrate that our proposed approach significantly improves the detection performance.

As one of the major branches of automatic speech recognition, attention-based models greatly improves the feature representation ability of the model. In particular, the multi-head mechanism is employed in the attention, hoping to learn speech features of more aspects in different attention subspaces. For speech recognition of complex languages, on the one hand, a small head size will lead to an obvious shortage of learnable aspects. On the other hand, we need to reduce the dimension of each subspace to keep the size of the overall feature space unchanged when we increase the number of heads, which will significantly weaken the ability to represent the feature of each subspace. Therefore, this paper explores how to use a small attention subspace to represent complete speech features while ensuring many heads. In this work we propose a novel neural network architecture, namely, pyramid multi-branch fusion DCNN with multi-head self-attention. The proposed architecture is inspired by Dilated Convolution Neural Networks (DCNN), it uses multiple branches with DCNN to extract the feature of the input speech under different receptive fields. To reduce the number of parameters, every two branches are merged until all the branches are merged into one. Thus, its shape is like a pyramid rotated 90 degrees. We demonstrate that on Aishell-1, a widely used Mandarin speech dataset, our model achieves a character error rate (CER) of 6.45% on the test sets.

We investigate the stability and robustness properties of a continuification-based strategy for the control of large-scale multiagent systems. Within continuation-based strategy, one transforms the microscopic, agent-level description of the system dynamics into a macroscopic continuum-level, for which a control action can be synthesized to steer the macroscopic dynamics towards a desired distribution. Such an action is ultimately discretized to obtain a set of deployable control inputs for the agents to achieve the goal. The mathematical proof of convergence toward the desired distribution typically relies on the assumptions that no disturbance is present and that each agent possesses global knowledge of all the others' positions. Here, we analytically and numerically address the possibility of relaxing these assumptions for the case of a one-dimensional system of agents moving in a ring. We offer compelling evidence in favor of the use of a continuification-based strategy when agents only possess a finite sensing capability and spatio-temporal perturbations affect the macroscopic dynamics of the ensemble. We also discuss some preliminary results about the role of an integral action in the macroscopic control solution.

Elongated anisotropic Gaussian filters are used for the orientation estimation of fibers. In cases where computed tomography images are noisy, roughly resolved, and of low contrast, they are the method of choice even if being efficient only in virtual 2D slices. However, minor inaccuracies in the anisotropic Gaussian filters can carry over to the orientation estimation. Therefore, we propose a modified algorithm for 2D anisotropic Gaussian filters and show that this improves their precision. Applied to synthetic images of fiber bundles, it is more accurate and robust to noise. Finally, we demonstrate the effectiveness of our approach by applying it to real-world images of sheet molding compounds.

Background and Objectives: Cardiovascular magnetic resonance (CMR) imaging is a powerful modality in functional and anatomical assessment for various cardiovascular diseases. Sufficient image quality is essential to achieve proper diagnosis and treatment. A large number of medical images, the variety of imaging artefacts, and the workload of imaging centres are among the things that reveal the necessity of automatic image quality assessment (IQA). However, automated IQA requires access to bulk annotated datasets for training deep learning (DL) models. Labelling medical images is a tedious, costly and time-consuming process, which creates a fundamental challenge in proposing DL-based methods for medical applications. This study aims to present a new method for CMR IQA when there is limited access to annotated datasets. Methods: The proposed generalised deep meta-learning model can evaluate the quality by learning tasks in the prior stage and then fine-tuning the resulting model on a small labelled dataset of the desired tasks. This model was evaluated on the data of over 6,000 subjects from the UK Biobank for five defined tasks, including detecting respiratory motion, cardiac motion, Aliasing and Gibbs ringing artefacts and images without artefacts. Results: The results of extensive experiments show the superiority of the proposed model. Besides, comparing the model's accuracy with the domain adaptation model indicates a significant difference by using only 64 annotated images related to the desired tasks. Conclusion: The proposed model can identify unknown artefacts in images with acceptable accuracy, which makes it suitable for medical applications and quality assessment of large cohorts.

In this paper, we introduce a Variational Autoencoder (VAE) based training approach that can compress and decompress cancer pathology slides at a compression ratio of 1:512, which is better than the previously reported state of the art (SOTA) in the literature, while still maintaining accuracy in clinical validation tasks. The compression approach was tested on more common computer vision datasets such as CIFAR10, and we explore which image characteristics enable this compression ratio on cancer imaging data but not generic images. We generate and visualize embeddings from the compressed latent space and demonstrate how they are useful for clinical interpretation of data, and how in the future such latent embeddings can be used to accelerate search of clinical imaging data.

This paper proposes a damage detection method based on the train-induced responses of high-speed railway box girder. Under the coupling effects of bending and torsion, the traditional damage detection method based on the Euler beam theory cannot be applied. In this research, the box girder section is divided into different components based on the plate element analysis method. The strain responses were preprocessed based on Principal Component Analysis (PCA) method to remove the influence of train operation variation. The residual error of autoregressive (AR) model was used as a potential index of damage features. The optimal order of the model was determined based on Bayesian information criterion (BIC) criterion. Finally, the confidence boundary (CB) of damage features (DF) constituting outliers can be estimated by Gaussian inverse cumulative distribution function (ICDF). The numerical simulation results show that the proposed method in this paper can effectively identify, locate and quantify the damage, which verifies the accuracy of the proposed method. The proposed method effectively identifies the early damage of all components on the key section by using four strain sensors, and it is helpful for developing effective maintenance strategies for high-speed railway box girder.

Motivated by applications in unmanned aerial based ground penetrating radar for detecting buried landmines, we consider the problem of imaging small point like scatterers situated in a lossy medium below a random rough surface. Both the random rough surface and the absorption in the lossy medium significantly impede the target detection and imaging process. Using principal component analysis we effectively remove the reflection from the air-soil interface. We then use a modification of the classical synthetic aperture radar imaging functional to image the targets. This imaging method introduces a user-defined parameter, $\delta$, which scales the resolution by $\sqrt{\delta}$ allowing for target localization with sub wavelength accuracy. Numerical results in two dimensions illustrate the robustness of the approach for imaging multiple targets. However, the depth at which targets are detectable is limited due to the absorption in the lossy medium.

Hyperspectral imaging systems that use multispectral filter arrays (MSFA) capture only one spectral component in each pixel. Hyperspectral demosaicing is used to recover the non-measured components. While deep learning methods have shown promise in this area, they still suffer from several challenges, including limited modeling of non-local dependencies, lack of consideration of the periodic MSFA pattern that could be linked to periodic artifacts, and difficulty in recovering high-frequency details. To address these challenges, this paper proposes a novel de-mosaicing framework, the MSFA-frequency-aware Transformer network (FDM-Net). FDM-Net integrates a novel MSFA-frequency-aware multi-head self-attention mechanism (MaFormer) and a filter-based Fourier zero-padding method to reconstruct high pass components with greater difficulty and low pass components with relative ease, separately. The advantage of Maformer is that it can leverage the MSFA information and non-local dependencies present in the data. Additionally, we introduce a joint spatial and frequency loss to transfer MSFA information and enhance training on frequency components that are hard to recover. Our experimental results demonstrate that FDM-Net outperforms state-of-the-art methods with 6dB PSNR, and reconstructs high-fidelity details successfully.

Current endpointing (EP) solutions learn in a supervised framework, which does not allow the model to incorporate feedback and improve in an online setting. Also, it is a common practice to utilize costly grid-search to find the best configuration for an endpointing model. In this paper, we aim to provide a solution for adaptive endpointing by proposing an efficient method for choosing an optimal endpointing configuration given utterance-level audio features in an online setting, while avoiding hyperparameter grid-search. Our method does not require ground truth labels, and only uses online learning from reward signals without requiring annotated labels. Specifically, we propose a deep contextual multi-armed bandit-based approach, which combines the representational power of neural networks with the action exploration behavior of Thompson modeling algorithms. We compare our approach to several baselines, and show that our deep bandit models also succeed in reducing early cutoff errors while maintaining low latency.

In TV services, dialogue level personalization is key to meeting user preferences and needs. When dialogue and background sounds are not separately available from the production stage, Dialogue Separation (DS) can estimate them to enable personalization. DS was shown to provide clear benefits for the end user. Still, the estimated signals are not perfect, and some leakage can be introduced. This is undesired, especially during passages without dialogue. We propose to combine DS and Voice Activity Detection (VAD), both recently proposed for TV audio. When their combination suggests dialogue inactivity, background components leaking in the dialogue estimate are reassigned to the background estimate. A clear improvement of the audio quality is shown for dialogue-free signals, without performance drops when dialogue is active. A post-processed VAD estimate with improved detection accuracy is also generated. It is concluded that DS and VAD can improve each other and are better used together.

Fog computing allows computationally-heavy problems with tight time constraints to be solved even if end devices have limited computational resources and latency induced by cloud computing is too high. How can energy consumed by fog computing be saved while obeying latency constraints and considering not only computations but also transmission through wireless and wired links? This work examines the latency and energy consumption sources in fog networks and discusses models describing these costs for various technologies. Next, resource allocation strategies are discussed considering the various degrees of freedom available in such a complex system, and their influence on energy consumption and latency. Finally, a vision for a future distributed, AI-driven resources allocation strategy is presented and justified.

Tracking interpersonal distances is essential for real-time social distancing management and {\em ex-post} contact tracing to prevent spreads of contagious diseases. Bluetooth neighbor discovery has been employed for such purposes in combating COVID-19, but does not provide satisfactory spatiotemporal resolutions. This paper presents ImmTrack, a system that uses a millimeter wave radar and exploits the inertial measurement data from user-carried smartphones or wearables to track interpersonal distances. By matching the movement traces reconstructed from the radar and inertial data, the pseudo identities of the inertial data can be transferred to the radar sensing results in the global coordinate system. The re-identified, radar-sensed movement trajectories are then used to track interpersonal distances. In a broader sense, ImmTrack is the first system that fuses data from millimeter wave radar and inertial measurement units for simultaneous user tracking and re-identification. Evaluation with up to 27 people in various indoor/outdoor environments shows ImmTrack's decimeters-seconds spatiotemporal accuracy in contact tracing, which is similar to that of the privacy-intrusive camera surveillance and significantly outperforms the Bluetooth neighbor discovery approach.

The adoption process of innovative software-intensive technologies leverages complex trust concerns in different forms and shapes. Perceived safety plays a fundamental role in technology adoption, being especially crucial in the case of those innovative software-driven technologies characterized by a high degree of dynamism and unpredictability, like collaborating autonomous systems. These systems need to synchronize their maneuvers in order to collaboratively engage in reactions to unpredictable incoming hazardous situations. That is however only possible in the presence of mutual trust. In this paper, we propose an approach for machine-to-machine dynamic trust assessment for collaborating autonomous systems that supports trust-building based on the concept of dynamic safety assurance within the collaborative process among the software-intensive autonomous systems. In our approach, we leverage the concept of digital twins which are abstract models fed with real-time data used in the run-time dynamic exchange of information. The information exchange is performed through the execution of specialized models that embed the necessary safety properties. More particularly, we examine the possible role of the Digital Twins in machine-to-machine trust building and present their design in supporting dynamic trust assessment of autonomous drones. Ultimately, we present a proof of concept of direct and indirect trust assessment by employing the Digital Twin in a use case involving two autonomous collaborating drones.

Although artificial intelligence (AI) systems have been shown to improve the accuracy of initial melanoma diagnosis, the lack of transparency in how these systems identify melanoma poses severe obstacles to user acceptance. Explainable artificial intelligence (XAI) methods can help to increase transparency, but most XAI methods are unable to produce precisely located domain-specific explanations, making the explanations difficult to interpret. Moreover, the impact of XAI methods on dermatologists has not yet been evaluated. Extending on two existing classifiers, we developed an XAI system that produces text and region based explanations that are easily interpretable by dermatologists alongside its differential diagnoses of melanomas and nevi. To evaluate this system, we conducted a three-part reader study to assess its impact on clinicians' diagnostic accuracy, confidence, and trust in the XAI-support. We showed that our XAI's explanations were highly aligned with clinicians' explanations and that both the clinicians' trust in the support system and their confidence in their diagnoses were significantly increased when using our XAI compared to using a conventional AI system. The clinicians' diagnostic accuracy was numerically, albeit not significantly, increased. This work demonstrates that clinicians are willing to adopt such an XAI system, motivating their future use in the clinic.

This work proposes a method to compute the maximum value obtained by a state function along trajectories of a Delay Differential Equation (DDE). An example of this task is finding the maximum number of infected people in an epidemic model with a nonzero incubation period. The variables of this peak estimation problem include the stopping time and the original history (restricted to a class of admissible histories). The original nonconvex DDE peak estimation problem is approximated by an infinite-dimensional Linear Program (LP) in occupation measures, inspired by existing measure-based methods in peak estimation and optimal control. This LP is approximated from above by a sequence of Semidefinite Programs (SDPs) through the moment-Sum of Squares (SOS) hierarchy. Effectiveness of this scheme in providing peak estimates for DDEs is demonstrated with provided examples

Inertia Axes are involved in many techniques for image content measurement when involving information obtained from lines, angles, centroids... etc. We investigate, here, the estimation of the main axis of inertia of an object in the image. We identify the coincidence conditions of the Scale Space Radon Transform (SSRT) maximum and the inertia main axis. We show, that by choosing the appropriate scale parameter, it is possible to match the SSRT maximum and the main axis of inertia location and orientation of the embedded object in the image. Furthermore, an example of use case is presented where binary objects central symmetry computation is derived by means of SSRT projections and the axis of inertia orientation. To this end, some SSRT characteristics have been highlighted and exploited. The experimentations show the SSRT-based main axis of inertia computation effectiveness. Concerning the central symmetry, results are very satisfying as experimentations carried out on randomly created images dataset and existing datasets have permitted to divide successfully these images bases into centrally symmetric and non-centrally symmetric objects.

In recent years, Transformer-based models such as the Switch Transformer have achieved remarkable results in natural language processing tasks. However, these models are often too complex and require extensive pre-training, which limits their effectiveness for small clinical text classification tasks with limited data. In this study, we propose a simplified Switch Transformer framework and train it from scratch on a small French clinical text classification dataset at CHU Sainte-Justine hospital. Our results demonstrate that the simplified small-scale Transformer models outperform pre-trained BERT-based models, including DistillBERT, CamemBERT, FlauBERT, and FrALBERT. Additionally, using a mixture of expert mechanisms from the Switch Transformer helps capture diverse patterns; hence, the proposed approach achieves better results than a conventional Transformer with the self-attention mechanism. Finally, our proposed framework achieves an accuracy of 87\%, precision at 87\%, and recall at 85\%, compared to the third-best pre-trained BERT-based model, FlauBERT, which achieved an accuracy of 84\%, precision at 84\%, and recall at 84\%. However, Switch Transformers have limitations, including a generalization gap and sharp minima. We compare it with a multi-layer perceptron neural network for small French clinical narratives classification and show that the latter outperforms all other models.

Existing audio-visual event localization (AVE) handles manually trimmed videos with only a single instance in each of them. However, this setting is unrealistic as natural videos often contain numerous audio-visual events with different categories. To better adapt to real-life applications, in this paper we focus on the task of dense-localizing audio-visual events, which aims to jointly localize and recognize all audio-visual events occurring in an untrimmed video. The problem is challenging as it requires fine-grained audio-visual scene and context understanding. To tackle this problem, we introduce the first Untrimmed Audio-Visual (UnAV-100) dataset, which contains 10K untrimmed videos with over 30K audio-visual events. Each video has 2.8 audio-visual events on average, and the events are usually related to each other and might co-occur as in real-life scenes. Next, we formulate the task using a new learning-based framework, which is capable of fully integrating audio and visual modalities to localize audio-visual events with various lengths and capture dependencies between them in a single pass. Extensive experiments demonstrate the effectiveness of our method as well as the significance of multi-scale cross-modal perception and dependency modeling for this task.

In the era of industrial big data, prognostics and health management is essential to improve the prediction of future failures to minimize inventory, maintenance, and human costs. Used for the 2021 PHM Data Challenge, the new Commercial Modular Aero-Propulsion System Simulation dataset from NASA is an open-source benchmark containing simulated turbofan engine units flown under realistic flight conditions. Deep learning approaches implemented previously for this application attempt to predict the remaining useful life of the engine units, but have not utilized labeled failure mode information, impeding practical usage and explainability. To address these limitations, a new prognostics approach is formulated with a customized loss function to simultaneously predict the current health state, the eventual failing component(s), and the remaining useful life. The proposed method incorporates principal component analysis to orthogonalize statistical time-domain features, which are inputs into supervised regressors such as random forests, extreme random forests, XGBoost, and artificial neural networks. The highest performing algorithm, ANN-Flux, achieves AUROC and AUPR scores exceeding 0.95 for each classification. In addition, ANN-Flux reduces the remaining useful life RMSE by 38% for the same test split of the dataset compared to past work, with significantly less computational cost.

We introduce LMCodec, a causal neural speech codec that provides high quality audio at very low bitrates. The backbone of the system is a causal convolutional codec that encodes audio into a hierarchy of coarse-to-fine tokens using residual vector quantization. LMCodec trains a Transformer language model to predict the fine tokens from the coarse ones in a generative fashion, allowing for the transmission of fewer codes. A second Transformer predicts the uncertainty of the next codes given the past transmitted codes, and is used to perform conditional entropy coding. A MUSHRA subjective test was conducted and shows that the quality is comparable to reference codecs at higher bitrates. Example audio is available at https://mjenrungrot.github.io/chrome-media-audio-papers/publications/lmcodec.

Intermittent computing systems undergo frequent power failure, hindering necessary data sample capture or timely on-device computation. These missing samples and deadlines limit the potential usage of intermittent computing systems in many time-sensitive and fault-tolerant applications. However, a group/swarm of intermittent nodes may amalgamate to sense and process all the samples by taking turns in waking up and extending their collective on-time. However, coordinating a swarm of intermittent computing nodes requires frequent and power-hungry communication, often infeasible with limited energy. Though previous works have shown promises when all intermittent nodes have access to the same amount of energy to harvest, work has yet to be looked into scenarios when the available energy distribution is different for each node. The proposed AICS framework provides an amalgamated intermittent computing system where each node schedules its wake-up schedules based on the duty cycle without communication overhead. We propose one offline tailored duty cycle selection method (Prime-Co-Prime), which schedules wake-up and sleep cycles for each node based on the measured energy to harvest for each node and the prior knowledge or estimation regarding the relative energy distribution. However, when the energy is variable, the problem is formulated as a Decentralized-Partially Observable Markov Decision Process (Dec-POMDP). Each node uses a group of heuristics to solve the Dec-POMDP and schedule its wake-up cycle.

Human-machine interaction (HMI) and human-robot interaction (HRI) can assist structural monitoring and structural dynamics testing in the laboratory and field. In vibratory experimentation, one mode of generating vibration is to use electrodynamic exciters. Manual control is a common way of setting the input of the exciter by the operator. To measure the structural responses to these generated vibrations sensors are attached to the structure. These sensors can be deployed by repeatable robots with high endurance, which require on-the-fly control. If the interface between operators and the controls was augmented, then operators can visualize the experiments, exciter levels, and define robot input with a better awareness of the area of interest. Robots can provide better aid to humans if intelligent on-the-fly control of the robot is: (1) quantified and presented to the human; (2) conducted in real-time for human feedback informed by data. Information provided by the new interface would be used to change the control input based on their understanding of real-time parameters. This research proposes using Augmented Reality (AR) applications to provide humans with sensor feedback and control of actuators and robots. This method improves cognition by allowing the operator to maintain awareness of structures while adjusting conditions accordingly with the assistance of the new real-time interface. One interface application is developed to plot sensor data in addition to voltage, frequency, and duration controls for vibration generation. Two more applications are developed under similar framework, one to control the position of a mediating robot and one to control the frequency of the robot movement. This paper presents the proposed model for the new control loop and then compares the new approach with a traditional method by measuring time delay in control input and user efficiency.

Determining clinically relevant physiological states from multivariate time series data with missing values is essential for providing appropriate treatment for acute conditions such as Traumatic Brain Injury (TBI), respiratory failure, and heart failure. Utilizing non-temporal clustering or data imputation and aggregation techniques may lead to loss of valuable information and biased analyses. In our study, we apply the SLAC-Time algorithm, an innovative self-supervision-based approach that maintains data integrity by avoiding imputation or aggregation, offering a more useful representation of acute patient states. By using SLAC-Time to cluster data in a large research dataset, we identified three distinct TBI physiological states and their specific feature profiles. We employed various clustering evaluation metrics and incorporated input from a clinical domain expert to validate and interpret the identified physiological states. Further, we discovered how specific clinical events and interventions can influence patient states and state transitions.

Pedestrian occlusion is challenging for autonomous vehicles (AVs) at midblock locations on multilane roadways because an AV cannot detect crossing pedestrians that are fully occluded by downstream vehicles in adjacent lanes. This paper tests the capability of vehicle-to-vehicle (V2V) communication between an AV and its downstream vehicles to share midblock pedestrian crossings information. The researchers developed a V2V-based collision-avoidance decision strategy and compared it to a base scenario (i.e., decision strategy without the utilization of V2V). Simulation results showed that for the base scenario, the near-zero time-to-collision (TTC) indicated no time for the AV to take appropriate action and resulted in dramatic braking followed by collisions. But the V2V-based collision-avoidance decision strategy allowed for a proportional braking approach to increase the TTC allowing the pedestrian to cross safely. To conclude, the V2V-based collision-avoidance decision strategy has higher safety benefits for an AV interacting with fully occluded pedestrians at midblock locations on multilane roadways.

Reconfigurable intelligent surface (RIS) has aroused a surge of interest in recent years. In this paper, we investigate the joint phase alignment and phase quantization on discrete phase shift designs for RIS-assisted single-input single-output (SISO) system. Firstly, the phenomena of phase distribution in far field and near field are respectively unveiled, paving the way for discretization of phase shift for RIS. Then, aiming at aligning phases, the phase distribution law and its underlying degree-of-freedom (DoF) are characterized, serving as the guideline of phase quantization strategies. Subsequently, two phase quantization methods, dynamic threshold phase quantization (DTPQ) and equal interval phase quantization (EIPQ), are proposed to strengthen the beamforming effect of RIS. DTPQ is capable of calculating the optimal discrete phase shifts with linear complexity in the number of unit cells on RIS, whilst EIPQ is a simplified method with a constant complexity yielding sub-optimal solution. Simulation results demonstrate that both methods achieve substantial improvements on power gain, stability, and robustness over traditional quantization methods. The path loss (PL) scaling law under discrete phase shift of RIS is unveiled for the first time, with the phase shifts designed by DTPQ due to its optimality. Additionally, the field trials conducted at 2.6 GHz and 35 GHz validate the favourable performance of the proposed methods in practical communication environment.

Transformer-based models have recently made significant achievements in the application of end-to-end (E2E) automatic speech recognition (ASR). It is possible to deploy the E2E ASR system on smart devices with the help of Transformer-based models. While these models still have the disadvantage of requiring a large number of model parameters. To overcome the drawback of universal Transformer models for the application of ASR on edge devices, we propose a solution that can reuse the block in Transformer models for the occasion of the small footprint ASR system, which meets the objective of accommodating resource limitations without compromising recognition accuracy. Specifically, we design a novel block-reusing strategy for speech Transformer (BRST) to enhance the effectiveness of parameters and propose an adapter module (ADM) that can produce a compact and adaptable model with only a few additional trainable parameters accompanying each reusing block. We conducted an experiment with the proposed method on the public AISHELL-1 corpus, and the results show that the proposed approach achieves the character error rate (CER) of 9.3%/6.63% with only 7.6M/8.3M parameters without and with the ADM, respectively. In addition, we also make a deeper analysis to show the effect of ADM in the general block-reusing method.

This work is unique in the use of discrete wavelets that were built from or derived from Chebyshev polynomials of the second and third kind, filter the Discrete Second Chebyshev Wavelets Transform (DSCWT), and derive two effective filters. The Filter Discrete Third Chebyshev Wavelets Transform (FDTCWT) is used in the process of analyzing color images and removing noise and impurities that accompany the image, as well as because of the large amount of data that makes up the image as it is taken. These data are massive, making it difficult to deal with each other during transmission. However to address this issue, the image compression technique is used, with the image not losing information due to the readings that were obtained, and the results were satisfactory. Mean Square Error (MSE), Peak Signal Noise Ratio (PSNR), Bit Per Pixel (BPP), and Compression Ratio (CR) Coronavirus is the initial treatment, while the processing stage is done with network training for Convolutional Neural Networks (CNN) with Discrete Second Chebeshev Wavelets Convolutional Neural Network (DSCWCNN) and Discrete Third Chebeshev Wavelets Convolutional Neural Network (DTCWCNN) to create an efficient algorithm for face recognition, and the best results were achieved in accuracy and in the least amount of time. Two samples of color images that were made or implemented were used. The proposed theory was obtained with fast and good results; the results are evident shown in the tables below.

The Internet has turned the entire world into a small village;this is because it has made it possible to share millions of images and videos. However, sending and receiving a huge amount of data is considered to be a main challenge. To address this issue, a new algorithm is required to reduce image bits and represent the data in a compressed form. Nevertheless, image compression is an important application for transferring large files and images. This requires appropriate and efficient transfers in this field to achieve the task and reach the best results. In this work, we propose a new algorithm based on discrete Hermite wavelets transformation (DHWT) that shows the efficiency and quality of the color images. By compressing the color image, this method analyzes it and divides it into approximate coefficients and detail coefficients after adding the wavelets into MATLAB. With Multi-Resolution Analyses (MRA), the appropriate filter is derived, and the mathematical aspects prove to be validated by testing a new filter and performing its operation. After the decomposition of the rows and upon the process of the reconstruction, taking the inverse of the filter and dealing with the columns of the matrix, the original matrix is improved by measuring the parameters of the image to achieve the best quality of the resulting image, such as the peak signal-to-noise ratio (PSNR), compression ratio (CR), bits per pixel (BPP), and mean square error (MSE).

Universal anomaly detection still remains a challenging problem in machine learning and medical image analysis. It is possible to learn an expected distribution from a single class of normative samples, e.g., through epistemic uncertainty estimates, auto-encoding models, or from synthetic anomalies in a self-supervised way. The performance of self-supervised anomaly detection approaches is still inferior compared to methods that use examples from known unknown classes to shape the decision boundary. However, outlier exposure methods often do not identify unknown unknowns. Here we discuss an improved self-supervised single-class training strategy that supports the approximation of probabilistic inference with loosen feature locality constraints. We show that up-scaling of gradients with histogram-equalised images is beneficial for recently proposed self-supervision tasks. Our method is integrated into several out-of-distribution (OOD) detection models and we show evidence that our method outperforms the state-of-the-art on various benchmark datasets. Source code will be publicly available by the time of the conference.

Machine learning algorithms, especially Neural Networks (NNs), are a valuable tool used to approximate non-linear relationships, like the AC-Optimal Power Flow (AC-OPF), with considerable accuracy -- and achieving a speedup of several orders of magnitude when deployed for use. Often in power systems literature, the NNs are trained with a fixed dataset generated prior to the training process. In this paper, we show that adapting the NN training dataset during training can improve the NN performance and substantially reduce its worst-case violations. This paper proposes an algorithm that identifies and enriches the training dataset with critical datapoints that reduce the worst-case violations and deliver a neural network with improved worst-case performance guarantees. We demonstrate the performance of our algorithm in four test power systems, ranging from 39-buses to 162-buses.

We consider high-dimensional MIMO transmissions in frequency division duplexing (FDD) systems. For precoding, the frequency selective channel has to be measured, quantized and fed back to the base station by the users. When the number of antennas is very high this typically leads to prohibitively high quantization complexity and large feedback. In 5G New Radio (NR), a modular quantization approach has been applied for this, where first a low-dimensional subspace is identified for the whole frequency selective channel, and then subband channels are linearly mapped to this subspace and quantized. We analyze how the components in such a modular scheme contribute to the overall quantization distortion. Based on this analysis we improve the technology components in the modular approach and propose an orthonormalized wideband precoding scheme and a sequential wideband precoding approach which provide considerable gains over the conventional method. We compare the performance of the developed quantization schemes to prior art by simulations in terms of the projection distortion, overall distortion and spectral efficiency, in a scenario with a realistic spatial channel model.

Instrument playing technique (IPT) is a key element of musical presentation. However, most of the existing works for IPT detection only concern monophonic music signals, yet little has been done to detect IPTs in polyphonic instrumental solo pieces with overlapping IPTs or mixed IPTs. In this paper, we formulate it as a frame-level multi-label classification problem and apply it to Guzheng, a Chinese plucked string instrument. We create a new dataset, Guzheng\_Tech99, containing Guzheng recordings and onset, offset, pitch, IPT annotations of each note. Because different IPTs vary a lot in their lengths, we propose a new method to solve this problem using multi-scale network and self-attention. The multi-scale network extracts features from different scales, and the self-attention mechanism applied to the feature maps at the coarsest scale further enhances the long-range feature extraction. Our approach outperforms existing works by a large margin, indicating its effectiveness in IPT detection.

The daily operation of real-world power systems and their underlying markets relies on the timely solution of the unit commitment problem. However, given its computational complexity, several optimization-based methods have been proposed to lighten its problem formulation by removing redundant line flow constraints. These approaches often ignore the spatial couplings of renewable generation and demand, which have an inherent impact of market outcomes. Moreover, the elimination procedures primarily focus on the feasible region and exclude how the problem's objective function plays a role here. To address these pitfalls, we move to rule out redundant and inactive constraints over a tight linear programming relaxation of the original unit commitment feasibility region by adding valid inequality constraints. We extend the optimization-based approach called umbrella constraint discovery through the enforcement of a consistency logic on the set of constraints by adding the proposed inequality constraints to the formulation. Hence, we reduce the conservativeness of the screening approach using the available historical data and thus lead to a tighter unit commitment formulation. Numerical tests are performed on standard IEEE test networks to substantiate the effectiveness of the proposed approach.

Generative AI has demonstrated impressive performance in various fields, among which speech synthesis is an interesting direction. With the diffusion model as the most popular generative model, numerous works have attempted two active tasks: text to speech and speech enhancement. This work conducts a survey on audio diffusion model, which is complementary to existing surveys that either lack the recent progress of diffusion-based speech synthesis or highlight an overall picture of applying diffusion model in multiple fields. Specifically, this work first briefly introduces the background of audio and diffusion model. As for the text-to-speech task, we divide the methods into three categories based on the stage where diffusion model is adopted: acoustic model, vocoder and end-to-end framework. Moreover, we categorize various speech enhancement tasks by either certain signals are removed or added into the input speech. Comparisons of experimental results and discussions are also covered in this survey.

Within academia and industry, there has been a need for expansive simulation frameworks that include model-based simulation of sensors, mobile vehicles, and the environment around them. To this end, the modular, real-time, and open-source AirSim framework has been a popular community-built system that fulfills some of those needs. However, the framework required adding systems to serve some complex industrial applications, including designing and testing new sensor modalities, Simultaneous Localization And Mapping (SLAM), autonomous navigation algorithms, and transfer learning with machine learning models. In this work, we discuss the modification and additions to our open-source version of the AirSim simulation framework, including new sensor modalities, vehicle types, and methods to generate realistic environments with changeable objects procedurally. Furthermore, we show the various applications and use cases the framework can serve.

Planning and control for uncertain contact systems is challenging as it is not clear how to propagate uncertainty for planning. Contact-rich tasks can be modeled efficiently using complementarity constraints among other techniques. In this paper, we present a stochastic optimization technique with chance constraints for systems with stochastic complementarity constraints. We use a particle filter-based approach to propagate moments for stochastic complementarity system. To circumvent the issues of open-loop chance constrained planning, we propose a contact-aware controller for covariance steering of the complementarity system. Our optimization problem is formulated as Non-Linear Programming (NLP) using bilevel optimization. We present an important-particle algorithm for numerical efficiency for the underlying control problem. We verify that our contact-aware closed-loop controller is able to steer the covariance of the states under stochastic contact-rich tasks.

Diffusion-based models for text-to-image generation have gained immense popularity due to recent advancements in efficiency, accessibility, and quality. Although it is becoming increasingly feasible to perform inference with these systems using consumer-grade GPUs, training them from scratch still requires access to large datasets and significant computational resources. In the case of medical image generation, the availability of large, publicly accessible datasets that include text reports is limited due to legal and ethical concerns. While training a diffusion model on a private dataset may address this issue, it is not always feasible for institutions lacking the necessary computational resources. This work demonstrates that pre-trained Stable Diffusion models, originally trained on natural images, can be adapted to various medical imaging modalities by training text embeddings with textual inversion. In this study, we conducted experiments using medical datasets comprising only 100 samples from three medical modalities. Embeddings were trained in a matter of hours, while still retaining diagnostic relevance in image generation. Experiments were designed to achieve several objectives. Firstly, we fine-tuned the training and inference processes of textual inversion, revealing that larger embeddings and more examples are required. Secondly, we validated our approach by demonstrating a 2\% increase in the diagnostic accuracy (AUC) for detecting prostate cancer on MRI, which is a challenging multi-modal imaging modality, from 0.78 to 0.80. Thirdly, we performed simulations by interpolating between healthy and diseased states, combining multiple pathologies, and inpainting to show embedding flexibility and control of disease appearance. Finally, the embeddings trained in this study are small (less than 1 MB), which facilitates easy sharing of medical data with reduced privacy concerns.

This paper describes our submission to ICASSP 2023 MUG Challenge Track 4, Keyphrase Extraction, which aims to extract keyphrases most relevant to the conference theme from conference materials. We model the challenge as a single-class Named Entity Recognition task and developed techniques for better performance on the challenge: For the data preprocessing, we encode the split keyphrases after word segmentation. In addition, we increase the amount of input information that the model can accept at one time by fusing multiple preprocessed sentences into one segment. We replace the loss function with the multi-class focal loss to address the sparseness of keyphrases. Besides, we score each appearance of keyphrases and add an extra output layer to fit the score to rank keyphrases. Exhaustive evaluations are performed to find the best combination of the word segmentation tool, the pre-trained embedding model, and the corresponding hyperparameters. With these proposals, we scored 45.04 on the final test set.

Humans naturally perceive surrounding scenes by unifying sound and sight in a first-person view. Likewise, machines are advanced to approach human intelligence by learning with multisensory inputs from an egocentric perspective. In this paper, we explore the challenging egocentric audio-visual object localization task and observe that 1) egomotion commonly exists in first-person recordings, even within a short duration; 2) The out-of-view sound components can be created while wearers shift their attention. To address the first problem, we propose a geometry-aware temporal aggregation module to handle the egomotion explicitly. The effect of egomotion is mitigated by estimating the temporal geometry transformation and exploiting it to update visual representations. Moreover, we propose a cascaded feature enhancement module to tackle the second issue. It improves cross-modal localization robustness by disentangling visually-indicated audio representation. During training, we take advantage of the naturally available audio-visual temporal synchronization as the ``free'' self-supervision to avoid costly labeling. We also annotate and create the Epic Sounding Object dataset for evaluation purposes. Extensive experiments show that our method achieves state-of-the-art localization performance in egocentric videos and can be generalized to diverse audio-visual scenes.

We consider a multi-process remote estimation system observing $K$ independent Ornstein-Uhlenbeck processes. In this system, a shared sensor samples the $K$ processes in such a way that the long-term average sum mean square error (MSE) is minimized. The sensor operates under a total sampling frequency constraint $f_{\max}$. The samples from all processes consume random processing delays in a shared queue and then are transmitted over an erasure channel with probability $\epsilon$. We study two variants of the problem: first, when the samples are scheduled according to a Maximum-Age-First (MAF) policy, and the receiver provides an erasure status feedback; and second, when samples are scheduled according to a Round-Robin (RR) policy, when there is no erasure status feedback from the receiver. Aided by optimal structural results, we show that the optimal sampling policy for both settings, under some conditions, is a \emph{threshold policy}. We characterize the optimal threshold and the corresponding optimal long-term average sum MSE as a function of $K$, $f_{\max}$, $\epsilon$, and the statistical properties of the observed processes. Our results show that, with an exponentially distributed service rate, the optimal threshold $\tau^*$ increases as the number of processes $K$ increases, for both settings. Additionally, we show that the optimal threshold is an \emph{increasing} function of $\epsilon$ in the case of \emph{available} erasure status feedback, while it exhibits the \emph{opposite behavior}, i.e., $\tau^*$ is a \emph{decreasing} function of $\epsilon$, in the case of \emph{absent} erasure status feedback.