Papers

No-regret learning dynamics play a central role in game theory, enabling decentralized convergence to equilibrium for concepts such as Coarse Correlated Equilibrium (CCE) or Correlated Equilibrium (CE). In this work, we improve the convergence rate to CCE in general-sum Markov games, reducing it from the previously best-known rate of $\mathcal{O}(\log^5 T / T)$ to a sharper $\mathcal{O}(\log T / T)$. This matches the best known convergence rate for CE in terms of $T$, number of iterations, while also improving the dependence on the action set size from polynomial to polylogarithmic-yielding exponential gains in high-dimensional settings. Our approach builds on recent advances in adaptive step-size techniques for no-regret algorithms in normal-form games, and extends them to the Markovian setting via a stage-wise scheme that adjusts learning rates based on real-time feedback. We frame policy updates as an instance of Optimistic Follow-the-Regularized-Leader (OFTRL), customized for value-iteration-based learning. The resulting self-play algorithm achieves, to our knowledge, the fastest known convergence rate to CCE in Markov games.

Collaborating teams of robots show promise due to their ability to complete missions more efficiently and with improved robustness, attributes that are particularly useful for systems operating in marine environments. A key issue is how to model, analyze, and design these multi-robot systems to realize the full benefits of collaboration, a challenging task since the domain of multi-robot autonomy encompasses both collective and individual behaviors. This paper introduces a layered model of multi-robot autonomy that uses the principle of census, or a weighted count of the inputs from neighbors, for collective decision-making about teaming, coupled with multi-objective behavior optimization for individual decision-making about actions. The census component is expressed as a nonlinear opinion dynamics model and the multi-objective behavior optimization is accomplished using interval programming. This model can be reduced to recover foundational algorithms in distributed optimization and control, while the full model enables new types of collective behaviors that are useful in real-world scenarios. To illustrate these points, a new method for distributed optimization of subgroup allocation is introduced where robots use a gradient descent algorithm to minimize portions of the cost functions that are locally known, while being influenced by the opinion states from neighbors to account for the unobserved costs. With this method the group can collectively use the information contained in the Hessian matrix of the total global cost. The utility of this model is experimentally validated in three categorically different experiments with fleets of autonomous surface vehicles: an adaptive sampling scenario, a high value unit protection scenario, and a competitive game of capture the flag.

The electric grid is increasingly vital, supporting essential services such as healthcare, heating and cooling transportation, telecommunications, and water systems. This growing dependence on reliable power underscores the need for enhanced grid resilience. This study presents Eversource's Climate Vulnerability Assessment (CVA) for bulk distribution substations in Massachusetts, evaluating risks from storm surge, sea level rise, precipitation, and extreme temperatures. The focus is on developing a cost-efficient model to guide targeted resilience investments. This is achieved by overcoming the limitations of single-variable analyses through hazard-specific assessments that integrate spatial, climate, electrical asset, and other relevant data; and applying sensitivity analysis to establish data-driven thresholds for actionable climate risks. By integrating geospatial analysis and data modeling with power engineering principles, this study provides a practical and replicable framework for equitable, data-informed climate adaptation planning. The results indicate that thresholds for certain climate hazards can be highly sensitive and result in significantly larger sets of stations requiring mitigation measures to adequately adapt to climate change, indicating that high-fidelity long-term climate projections are critical.

We propose a Model Predictive Control (MPC) formulation for nonlinear systems without terminal penalty or dedicated stabilizing terminal set, in which state constraints are enforced heterogeneously along the prediction horizon. In this setting a control-invariant set certifies near-term safety, while a less restrictive set constrains later predictions. This structure is motivated by safety-critical formulations, such as Control Barrier Function (CBF) based MPC, collision avoidance, and robotic receding-horizon planning, where near-term predictions must be certified safe and later predictions can be re-certified in future updates. We develop a value-function-difference analysis separating the effects of constraint-set selection and prediction-horizon length on closed-loop performance bounds, yielding implicit suboptimality certificates. Assuming cost-controllability, we further propose an upper-bound certificate which accounts explicitly for distinct constraint sets, their associated decay rates, and horizons. We also provide a lower-bound certificate for the closed-loop cost beyond the finite-horizon open-loop cost, which, for the admissible parameter range, is not weaker than the standard finite-horizon bound. Simulations on linear and nonlinear safety-critical systems demonstrate the proposed certificates a priori and a posteriori.

This paper proposes two projector-based Hopfield neural network (HNN) estimators for online, constrained parameter estimation under time-varying data, additive disturbances, and slowly drifting physical parameters. The first is a constraint-aware HNN that enforces linear equalities and inequalities (via slack neurons) and continuously tracks the constrained least-squares target. The second augments the state with compensation neurons and a concatenated regressor to absorb bias-like disturbance components within the same energy function. For both estimators we establish global uniform ultimate boundedness with explicit convergence rate and ultimate bound, and we derive practical tuning rules that link the three design gains to closed-loop bandwidth and steady-state accuracy. We also introduce an online identifiability monitor that adapts the constraint weight and time step, and, when needed, projects updates onto identifiable subspaces to prevent drift in poorly excited directions...

Molecular communication (MC) is a promising paradigm for applications where traditional electromagnetic communications are impractical. However, decoding chemical signals, especially in multi-transmitter systems, remains a key challenge due to interference and complex propagation dynamics. In this paper, we develop a one-dimensional fractal convolutional neural network (fCNN) to detect the concentrations of multiple types of molecules based on the absorbance spectra measured at a receiver. Our model is trained by both experimental and simulated datasets, with the latter enhanced by noise modeling to mimic real-world measurements. We demonstrate that a noiseaugmented simulated dataset can effectively be a substitute for experimental data, achieving similar decoding accuracy. Our approach successfully detects bit sequences in both binary and quadruple concentration shift keying (BCSK and QCSK) scenarios, even when transmitters are desynchronized, highlighting the potential of machine learning for robust MC signal detection.

Prosody is essential for speech technology, shaping comprehension, naturalness, and expressiveness. However, current text-to-speech (TTS) systems still struggle to accurately capture human-like prosodic variation, in part because existing evaluation methods for prosody remain limited. Traditional metrics like Mean Opinion Score (MOS) are resource-intensive, inconsistent, and offer little insight into why a system sounds unnatural. This study introduces a linguistically informed, semi-automatic framework for evaluating TTS prosody through a two-tier architecture that mirrors human prosodic organization. The method uses quantitative linguistic criteria to evaluate synthesized speech against human speech corpora across multiple acoustic dimensions. By integrating discrete and continuous prosodic measures, it provides objective and interpretable metrics of both event placement and cue realization, while accounting for the natural variability observed across speakers and prosodic cues. Results show strong correlations with perceptual MOS ratings while revealing model-specific weaknesses that traditional perceptual tests alone cannot capture. This approach provides a principled path toward diagnosing, benchmarking, and ultimately improving the prosodic naturalness of next-generation TTS systems.

Policy gradient methods are a powerful family of reinforcement learning algorithms for continuous control that optimize a policy directly. However, standard first-order methods often converge slowly. Second-order methods can accelerate learning by using curvature information, but they are typically expensive to compute. The linear quadratic regulator (LQR) is a practical setting in which key quantities, such as the policy gradient, admit closed-form expressions. In this work, we develop second-order policy gradient algorithms for LQR by deriving explicit formulas for both the approximate and exact Hessians used in Gauss–Newton and Newton methods, respectively. Numerical experiments show a faster convergence rate for the proposed second-order approach over the standard first-order policy gradient baseline.

Conventional relays encounter difficulties in protecting transmission lines (TLs) connected to converter-based energy sources (CBESs) due to the influence of power electronics on fault characteristics. This article proposes a single-ended intelligent protection method for the TL segment between the grid and a Photovoltaic (PV) plant. The approach utilizes a Recurrence Matrix and an InceptionTime-based system to identify faults by using the mean change in quantiles of 3-phase currents. It determines the fault position and identifies the faulty phase. ReliefF feature selection is applied to extract the optimal quantile features. The scheme's performance is assessed under abnormal conditions, including faults and capacitor and load-switching events, simulated in Power Systems Computer Aided Design / Electromagnetic Transients Program (PSCAD/EMTDC) on the Western System Coordinating Council (WSCC) 9-bus system, with various fault and switching parameters. The scheme is also validated on the New England IEEE 39-bus system and in presence of partially rated converters. Additionally, the validation of the proposed strategy takes into account various conditions, including double-circuit line configuration, noise, series compensation, high-impedance faults, current transformer (CT) saturation, evolving and cross-country faults, remote and local faults, as well as variations in PV capacity, sampling frequency, and data window size. To address label scarcity and improve generalization, semi-supervised learning paradigms including label spreading, label propagation, and self-training are integrated with the InceptionTime framework, enabling near-supervised performance with limited annotated fault data. The results demonstrate that the approach is effective in handling different system configurations and conditions, ensuring the protection of TLs connected to large PV plants.

Molecular Communications (MC) is an emerging research paradigm that utilizes molecules to transmit information, with promising applications in biomedicine such as targeted drug delivery or tumor detection. It is also envisioned as a key enabler of the Internet of BioNanoThings (IoBNT). In this paper, we propose algorithms based on Recurrent Neural Networks (RNN) for the estimation of communication channel parameters in MC systems. We focus on a simple branched topology, simulating the molecule movement with a macroscopic MC simulator. The Deep Learning architectures proposed for distance estimation demonstrate strong performance within these branched environments, highlighting their potential for future MC applications.

Hybrid opto-electronic neural networks combine optical front-ends with electronic back-ends to perform vision tasks, but joint end-to-end (E2E) optimization of optical and electronic components is computationally expensive due to large parameter spaces and repeated optical convolutions. We propose Direct Kernel Optimization (DKO), a two-stage training framework that first trains a conventional electronic CNN and then synthesizes optical kernels to replicate the first-layer convolutional filters, reducing optimization dimensionality and avoiding hefty simulated optical convolutions during optimization. We evaluate DKO in simulation on a monocular depth estimation model and show that it achieves twice the accuracy of E2E training under equal computational budgets while reducing training time. Given the substantial computational challenges of optimizing hybrid opto-electronic systems, our results position DKO as a scalable optimization approach to train and realize these systems.

Due to the rapid growth of heterogeneous wireless networks (HWNs), where devices with diverse communication technologies coexist, there is increasing demand for efficient and adaptive multi-hop routing with multiple data flows. Traditional routing methods, designed for homogeneous environments, fail to address the complexity introduced by links consisting of multiple technologies, frequency-dependent fading, and dynamic topology changes. In this paper, we propose a deep reinforcement learning (DRL)-based routing framework using deep Q-networks (DQN) to establish routes between multiple source-destination pairs in HWNs by enabling each node to jointly select a communication technology, a subband, and a next hop relay that maximizes the rate of the route. Our approach incorporates channel and interference-aware neighbor selection approaches to improve decision-making beyond conventional distance-based heuristics. We further evaluate the robustness and generalizability of the proposed method under varying network dynamics, including node mobility, changes in node density, and the number of data flows. Simulation results demonstrate that our DRL-based routing framework significantly enhances scalability, adaptability, and end-to-end throughput in complex HWN scenarios.

Classical optimal execution models treat market impact as a pre-specified, exogenous process. However, when market makers adapt strategically, this assumption becomes a structural misspecification: execution dynamics depend on the policies and actions of other agents. The problem therefore ceases to be a single-agent control problem and instead becomes a finite-horizon stochastic game, in which liquidity emerges endogenously from the interactions among heterogeneous market players. We hence introduce ABIDES-MARL, a multi-agent reinforcement learning framework for studying optimal execution under endogenous liquidity in a realistic limit order book setting. The framework extends ABIDES-Gym to support multiple learning agents with synchronized decision periods that preserve proper information filtration and key market microstructure features. We validate the framework in an extended Kyle setting, where multiple learning agents recovers gradual price discovery. Numerical results then show that execution strategies learned under endogenous, adaptive liquidity differ fundamentally from those implied by fixed exogenous price-impact benchmarks. In particular, strategies optimized under exogenous impact perform poorly once market makers adapt strategically: when information is not balanced across traders and market makers, market makers exploit predictable execution and the market dynamics may degenerate. The framework therefore provides a reproducible foundation for studying strategic adaptation in realistic markets and contributes to the development of economically interpretable agentic AI systems in finance.

Road grade can impact the energy efficiency, safety, and comfort associated with automated vehicle control systems. Currently, control systems that attempt to compensate for road grade are designed with one of two assumptions. Either the grade is only known once the vehicle is driving over the road segment through proprioception, or complete knowledge of the oncoming road grade is known from a pre-made map. Both assumptions limit the performance of a control system, as not having a preview signal prevents proactive grade compensation, whereas relying only on map data potentially subjects the control system to missing or outdated information. These limits can be avoided by measuring the oncoming grade in real-time using on-board lidar sensors. In this work, we use point returns accumulated during travel to estimate the grade at each waypoint along a path. The estimated grade is defined as the difference in height between the front and rear wheelbase at a given waypoint. Kalman filtering techniques are used to mitigate the effects of odometry and motion uncertainty on the grade estimates. This estimator's performance is compared to the measurements of a map created with a GNSS/INS system via a field experiment. When compared to the map-based system, the lidar-based estimator produces an unbiased error with a standard deviation of 0.6 degrees at an average range of 52.7 meters. By having similar precision to map-based systems, automotive lidar-based grade estimation systems are shown to be a valid approach for measuring road grade when a map is unavailable or inaccurate. In using lidar as an input signal for grade-based control system tasks, autonomous vehicles achieve higher redundancy and independence in contrast to existing methods.

Background and objective: Prior probability shift between training and deployment datasets challenges deep learning-based medical image classification. Standard correction methods reweight posterior probabilities to adjust prior bias, yet their benefit is inconsistent. We developed a reliability framework identifying when prior correction helps or harms performance in pathological cell image analysis. Methods: We analyzed 303 colorectal cancer specimens with CD103/CD8 immunostaining, yielding 185,432 annotated cell images across 16 cell types. ResNet models were trained under varying bias ratios (1.1-20$\times$). Feature separability was quantified using cosine similarity-based likelihood quality scores, reflecting intra- versus inter-class distinctions in learned feature spaces. Multiple linear regression, ANOVA, and generalized additive models (GAMs) evaluated associations among feature separability, prior bias, sample adequacy, and F1 performance. Results: Feature separability dominated performance ($\beta = 1.650$, $p < 0.001$), showing 412-fold stronger impact than prior bias ($\beta = 0.004$, $p = 0.018$). GAM analysis showed strong predictive power ($R^2 = 0.876$) with mostly linear trends. A quality threshold of 0.294 effectively identified cases requiring correction (AUC = 0.610). Cell types scoring $>0.5$ were robust without correction, whereas those $<0.3$ consistently required adjustment. Conclusion: Feature extraction quality, not bias magnitude, governs correction benefit. The proposed framework provides quantitative guidance for selective correction, enabling efficient deployment and reliable diagnostic AI.

Cardiovascular disease (CVD) remains a critical global health concern, demanding reliable and interpretable predictive models for early risk assessment. This study presents a large-scale analysis using the Heart Disease Health Indicators Dataset, developing a strategically weighted ensemble model that combines tree-based methods (LightGBM, XGBoost) with a Convolutional Neural Network (CNN) to predict CVD risk. The model was trained on a preprocessed dataset of 229,781 patients where the inherent class imbalance was managed through strategic weighting and feature engineering enhanced the original 22 features to 25. The final ensemble achieves a statistically significant improvement over the best individual model, with a Test AUC of 0.8371 (p=0.003) and is particularly suited for screening with a high recall of 80.0%. To provide transparency and clinical interpretability, surrogate decision trees and SHapley Additive exPlanations (SHAP) are used. The proposed model delivers a combination of robust predictive performance and clinical transparency by blending diverse learning architectures and incorporating explainability through SHAP and surrogate decision trees, making it a strong candidate for real-world deployment in public health screening.

The surge in AI workloads and escalating data center requirements have created demand for ultra-high-speed interconnects exceeding 200 Gb/s. As unit intervals (UI) shrink, even a few picoseconds of intra-pair skew can significantly degrade serializer-deserializer (SerDes) performance. To quantify the impact of intra-pair skew, conventional time-domain methods are often unreliable for coupled interconnects due to skew variations across voltage levels, while frequency-domain approaches frequently fail to address reciprocity and symmetry issues. This can result in channels that meet skew specifications in one direction but not the other, despite the inherently reciprocal nature of skew impact. To address these limitations, we introduce two new reciprocal parameters for quantifying intra-pair skew effects: Skew-Induced Insertion Loss Deviation (SILD) and its complementary Figure of Merit (FOM SILD). Measurements conducted using 224 Gb/s SerDes IP and a variety of channels with different intra-pair skews demonstrate a strong correlation between FOM SILD and bit error rate (BER). Results show that when FOM SILD is below 0.2-0.3 dB, BER remains stable, indicating minimal signal integrity degradation; however, BER increases noticeably as FOM SILD exceeds 0.3 dB. Statistical analysis across more than 3,000 high-speed twinax cables reveals that the majority exhibit FOM SILD values less than 0.1 dB, underscoring the practical relevance of the proposed metrics for high-speed interconnect assessment.

Holographic multiple-input multiple-output (MIMO) has emerged as a key enabler for 6G networks, yet conventional planar implementations suffer from spatial correlation and mutual coupling at sub-wavelength spacing, which fundamentally limit the effective degrees of freedom (EDOF) and channel capacity. Three-dimensional (3-D) holographic MIMO offers a pathway to overcome these constraints by exploiting volumetric array configurations that enlarge the effective aperture and unlock additional spatial modes. This work presents the first systematic evaluation that jointly incorporates electromagnetic (EM) characteristics, such as mutual coupling and radiation efficiency, into the analysis of 3-D arrays under Clarke, Kronecker, and standardized 3rd Generation Partnership Project (3GPP) channel models. Analytical derivations and full-wave simulations demonstrate that 3-D architectures achieve higher EDOF, narrower beamwidths, and notable capacity improvements compared with planar baselines. In 3GPP urban macro channels with horizontal element spacing of 0.3 lambda, 3-D configurations yield approximately 20% capacity improvement over conventional 2-D arrays, confirming the robustness and scalability of volumetric designs under realistic conditions. These findings bridge the gap between theoretical feasibility and practical deployment, offering design guidance for next-generation 6G base station arrays.

This paper presents the MOBIUS platform, a bipedal robot capable of walking, crawling, climbing, and rolling. MOBIUS features four limbs, two 6-DoF arms with two-finger grippers for manipulation and climbing, and two 4-DoF legs for locomotion–enabling smooth transitions across diverse terrains without reconfiguration. A hybrid control architecture combines reinforcement learning for locomotion and force control for compliant contact interactions during manipulation. A high-level MIQCP planner autonomously selects locomotion modes to balance stability and energy efficiency. Hardware experiments demonstrate robust gait transitions, dynamic climbing, and full-body load support via pinch grasp. Overall, MOBIUS demonstrates the importance of tight integration between morphology, high-level planning, and control to enable mobile loco-manipulation and grasping, substantially expanding its interaction capabilities, workspace, and traversability.

Photoplethysmography (PPG) is widely used as a non-invasive and accessible modality for continuous health monitoring. However, despite being a peripheral hemodynamic signal intrinsically coupled with systemic circulation, existing research has largely confined its scope to a narrow range of cardiovascular tasks, leaving a fundamental question underexplored: to what extent can PPG support holistic health profiling beyond traditional cardiovascular applications? To answer this question, we present AnyPPG, a foundation model-based framework designed to reveal the broader health-profiling potential of PPG. To ensure reliable performance for this investigation, AnyPPG is pretrained with ECG guidance on the most diverse PPG corpus with synchronized ECG to date, comprising over 100,000 hours of recordings from six large-scale data sources. This pretraining yields robust and physiologically grounded PPG representations that provide a reliable basis for subsequent analysis. Building upon this pretrained model, we conduct a systematic investigation into the association between PPG and holistic health through, to our knowledge, the first PPG-based phenome-wide disease detection study, spanning 1,468 disease phenotypes in more than 15,000 subjects. Our evaluation demonstrates the effectiveness of AnyPPG: across eight clinical and wearable datasets covering 15 downstream tasks, it achieves the best performance in 13 tasks. More importantly, in the phenome-wide analysis, AnyPPG exhibits meaningful discriminative capability (AUC $\ge$ 0.70) for 307 phenotypes across 16 distinct phecode chapters, including 230 non-circulatory conditions such as dementia and chronic kidney disease, many of which have rarely been explored using PPG. Collectively, these findings indicate that easily acquired PPG signals encode rich health-related information extending well beyond conventional cardiovascular assessment.

The growing demand for on-device large language model (LLM) inference highlights the need for efficient mobile edge computing (MEC) solutions, especially in resource-constrained settings. Speculative decoding offers a promising solution by partitioning token generation between a lightweight draft model on mobile devices and a powerful target model on edge servers, but suffers from communication overhead and asynchronous delays. This paper is the first to propose a unified framework that jointly optimizes user association and resource allocation (UARA) to support efficient parallel speculative decoding. We solve the UARA problem using a multi-agent deep reinforcement learning algorithm. To evaluate our approach under realistic conditions, we conduct experiments using the Sionna simulator. Results show that our method achieves up to 28.0% and an average of 23.7% reduction in end-to-end latency without compromising inference accuracy, enabling scalable and low-latency LLM services in MEC systems.

Target Language Extraction aims to extract speech in a specific language from a mixture waveform that contains multiple speakers speaking different languages. The human auditory system is adept at performing this task with the knowledge of the particular language. However, the performance of the conventional extraction systems is limited by the lack of this prior knowledge. Speech pre-trained models, which capture rich linguistic and phonetic representations from large-scale in-the-wild corpora, can provide this missing language knowledge to these systems. In this work, we propose a novel end-to-end framework to leverage language knowledge from speech pre-trained models. This knowledge is used to guide the extraction model to better capture the target language characteristics, thereby improving extraction quality. To demonstrate the effectiveness of our proposed approach, we construct the first publicly available multilingual dataset for Target Language Extraction. Experimental results show that our method achieves improvements of 1.22 dB and 1.12 dB in SI-SNR for English and German extraction, respectively, from mixtures containing both languages.

This paper investigates the idea of designing data-driven partial estimators for nonlinear systems showing parametric uncertainties using sparse multivariate polynomial relationships. A general framework is first presented and then validated on two illustrative examples with comparison to different possible Machine/Deep-Learning based alternatives. The results suggests the superiority of the proposed sparse identification scheme, at least when the learning data is small.

Image downscaling is a fundamental operation in image processing, crucial for adapting high-resolution content to various display and storage constraints. While classic methods often introduce blurring or aliasing, recent learning-based approaches offer improved adaptivity. However, achieving maximal fidelity against ground-truth low-resolution (LR) images, particularly by accounting for channel-specific characteristics, remains an open challenge. This paper introduces ADK-Net (Adaptive Downscaling Kernel Network), a novel deep convolutional neural network framework for high-fidelity supervised image downscaling. ADK-Net explicitly addresses channel interdependencies by learning to predict spatially-varying, adaptive resampling kernels independently for each pixel and uniquely for each color channel (RGB). The architecture employs a hierarchical design featuring a ResNet-based feature extractor and parallel channel-specific kernel generators, themselves composed of ResNet-based trunk and branch sub-modules, enabling fine-grained kernel prediction. Trained end-to-end using an L1 reconstruction loss against ground-truth LR data, ADK-Net effectively learns the target downscaling transformation. Extensive quantitative and qualitative experiments on standard benchmarks, including the RealSR dataset, demonstrate that ADK-Net establishes a new state-of-the-art in supervised image downscaling, yielding significant improvements in PSNR and SSIM metrics compared to existing learning-based and traditional methods.

The coexistence of radar and communications in wireless systems marks a paradigm shift for the sixth-generation (6G) networks. As 6G systems are expected to operate at higher frequencies and employ larger antenna arrays than fifth-generation (5G) systems, they can also enable more accurate sensing capabilities. To this end, the integrated sensing and communication (ISAC) paradigm aims to unify the physical and radio frequency (RF) domains by introducing the sensing functionality into the communication network. However, the clutter poses a challenge, as it can significantly degrade the sensing accuracy in ISAC systems. This paper presents a novel two-dimensional root multiple signal classification (2D-rootMUSIC)-based algorithm for static background clutter suppression. Computer simulation results indicate that the proposed method effectively mitigates the strong background clutter, yields accurate parameter estimation performance, and offers a notable improvement in the signal-to-clutter-and-noise ratio (SCNR), while outperforming the prior-art benchmark methods.

Movable antennas (MA) and transmissive reconfigurable intelligent surfaces (TRIS) represent two innovative technologies that significantly enhance the flexibility of wireless communication systems. In this paper, we propose a novel and compact base station architecture that synergistically integrates a movable antenna with a transmissive RIS in the near field, enabling joint optimization of antenna positioning and TRIS phase adjustments. The proposed model compensates for phase quantization loss and significantly enhances signal strength, even with low-resolution (1-2 bit) phase shifters. Leveraging this framework, we systematically evaluate system performance as a function of TRIS size and antenna placement. Our results indicate that antenna mobility provides an additional degree of freedom to enhance the desired signal and achieve a higher SNR, particularly when combined with TRIS capabilities. These findings demonstrate that MA-TRIS integration offers a cost-effective and energy-efficient pathway toward compact 6G base stations, combining hardware simplicity with strong performance gains.

Numerically efficient and stable algorithms are essential for kernel-based regularized system identification. The state of art algorithms exploit the semiseparable structure of the kernel and are based on the generator representation of the kernel matrix. However, as will be shown from both the theory and the practice, the algorithms based on the generator representation are sometimes numerically unstable, which limits their application in practice. This paper aims to address this issue by deriving and exploiting an alternative Givens-vector representation of some widely used kernel matrices. Based on the Givens-vector representation, we derive algorithms that yield more accurate results than existing algorithms without sacrificing efficiency. We demonstrate their usage for the kernel-based regularized system identification. Monte Carlo simulations show that the proposed algorithms admit the same order of computational complexity as the state-of-the-art ones based on generator representation, but without issues with numerical stability.

Large multiple antenna arrays coupled with accurate beamforming are essential in terahertz (THz) communications to ensure link reliability. However, as the number of antennas increases, beam alignment (focusing) and beam tracking in mobile networks incur prohibitive overhead. Additionally, the near-field region expands both with the size of antenna arrays and the carrier frequency, calling for adjustments in the beamforming to account for spherical wavefront instead of the conventional planar wave assumption. In this letter, we introduce a novel beam coherence time for mobile THz networks, to drastically reduce the rate of beam updates. Then, we propose a deep learning model, relying on a simple feedforward neural network with a time-dependent input, to predict the beam coherence time and adjust the beamforming on the fly with minimal overhead. Our numerical results demonstrate the effectiveness of the proposed approach by enabling higher data rates while reducing the overhead, especially at high (i.e., vehicular) mobility.

We introduce the notion of descriptor embedding for nonlinear systems and use it for the data-driven design of stabilizing controllers. Specifically, we provide sufficient data-dependent LMI conditions which, if feasible, return a stabilizing nonlinear controller of the form $u=K(x)Z(x)$ where $K(x)$ belongs to a polytope and $Z$ is a user-defined function. The proposed method is then extended to account for the presence of uncertainties and noisy data. Furthermore, a method to estimate the resulting region of attraction is given using only data. Simulation examples are used to illustrate the results and compare them to existing methods from the literature.

We study a dynamic game with a large population of players who choose actions from a finite set in continuous time. Each player has a state in a finite state space that evolves stochastically with their actions. A player's reward depends not only on their own state and action but also on the distribution of states and actions across the population, capturing effects such as congestion in traffic networks. While prior work in evolutionary game theory has primarily focused on static games without individual player state dynamics, we present the first comprehensive evolutionary analysis of such dynamic games. We propose an evolutionary model together with a mean field approximation of the finite-population game and establish strong approximation guarantees. We show that standard solution concepts for dynamic games lack an evolutionary interpretation, and we propose a new concept - the Mixed Stationary Nash Equilibrium (MSNE) - which admits one. We analyze the relationship between MSNE and the rest points of the mean field evolutionary model and study the evolutionary stability of MSNE.