Searcharxiv⌕ Search

arXiv subjects

Stefan Werner

Publications and source records attributed to Stefan Werner.

At least 19 recordsLinked to original sources

Physical Prior Mismatch in Variational Dual-Polarization Quantum SATCOM Channel Estimation

Dual-polarization quantum satellite communications require accurate estimation of complex polarization channels. We investigate physical latent-prior mismatch in variational estimation of a $2\times 2$ complex channel, using eight Cartesian latent coordinates and a fixed latent-to-channel mapping. Gaussian, matched and sharpened Rician, Laplace, and Cauchy priors are compared under a fixed neural architecture and forward Kullback-Leibler objective. Models are trained for 300,000 updates with $ζ=1/(2N_{\mathrm{est}})$ and evaluated using three training seeds and 10,000 shared held-out blocks per operating point. The Gaussian and Rician priors coincide at the Rayleigh limit. At higher Rician factors, the matched prior improves estimation accuracy, especially with fewer pilots. At $K=100$ and $N_{\mathrm{est}}=32$, the matched Rician variational estimator reduces mean squared Frobenius error by 2.50% relative to linear minimum mean-square error (LMMSE). We also derive a joint Rician Bayesian posterior-mean estimator that incorporates the complete least-squares noise covariance and reduces posterior integration to one-dimensional quadrature. It improves on LMMSE by 3.33% at the same operating point and achieves lower observed error than the scalar Rician and both neural estimators at all 12 operating points. Additional Rician prior sharpening increases error, while Laplace and Cauchy controls remain near the Gaussian baseline.

eess.SP↗

DPA: Decentralized Primal Averaging with Quasi-Global Momentum for Highly Heterogeneous Data

Decentralized training of deep learning models keeps training data local and supports on-device learning over networks. In realistic scenarios, heterogeneity across agents' local data poses an optimization challenge and can severely degrade test accuracy, especially on sparse communication graphs. Building on recent advances in centralized primal averaging, we propose Decentralized Primal Averaging (DPA), a decentralized learning algorithm that combines primal averaging with a local quasi-global momentum buffer. DPA maintains three coupled sequences for optimization, gradient evaluation, and output averaging, while using consecutive mixed iterates to construct a network-aware momentum direction without transmitting an auxiliary tracking variable. For smooth nonconvex objectives, we establish a convergence bound for DPA and identify conditions for linear speedup in the number of agents for sufficiently many iterations. We also introduce DPA-1G, a variant that uses one communication round per iteration and matches the per-iteration communication cost of DSGD-style baselines. Experiments following the GUT decentralized image-classification benchmark show that DPA compares favorably with DSGD, gradient tracking, QG-DSGDm, momentum tracking, and QG-GUTm across datasets, architectures, topologies, and heterogeneity levels, with the largest gains under high heterogeneity on sparse rings. At the same total communication cost, both DPA variants achieve higher mean test accuracy than the best baseline on CIFAR-10 and CIFAR-100 with ResNet-20. Our code is publicly available.

cs.LG↗

Low-Latency Joint Source-Channel Coding with Multiple Constant Curvature Curves

We propose constant-curvature curve multicode (C3M), a multi-branch extension of constant-curvature curve tube (C3T) analog joint source-channel coding for one-shot transmission over fixed isotropic AWGN channels. Instead of mapping a scalar source to a single curve on the power sphere, C3M partitions the source interval across a union of constant-speed branch curves in even-dimensional Euclidean space. This construction increases the available geometric degrees of freedom while preserving a low block length analog encoding structure. We derive the induced mixture output distribution, posterior representation, MAP and MMSE estimators, and mutual-information decomposition into branch-identification and along-curve components. We then introduce geometry- and classification-aware branch allocation rules based on tube density, curve length, and branch-classification reliability. The resulting design procedure fixes the optimal C3T curve as a starting branch and sequentially selects additional branches, thereby reducing the combinatorial cost of exhaustive multi-curve search. We use NORM-inspired branch-conditioned spectral decoders based on Fourier-feature lifting of the received ambient vector. Numerical experiments in four and six dimensions show that adding candidate branch curves improves the finite-dimensional end-to-end signal-to-distortion ratio, with selected two- and three-candidate C3M systems outperforming the one-curve baseline at +10 dB SNR. The results show that multi-curve analog mappings provide a practical finite-dimensional design route for low block length joint source-channel coding.

eess.SP↗

Variational Blind Equalization for Dual-Polarization Free-Space Quantum Communications

Dual-polarization free-space quantum communication links require receiver equalization under polarization mixing, fading, phase variation, and limited channel supervision. This paper investigates variational blind and semi-blind equalization using VISTA (Variational Inference for Signal Tracking and Adaptation), a variational autoencoder-based probabilistic physical parameter estimator. In the semi-blind branch, channel-relevant statistics are mapped to the underlying 2x2 complex polarization channel. We compare forward KL, reverse KL, Average Jeffreys, Max-KL, and Adaptive Jeffreys regularization and analyze directional KL asymmetry and the local effect of the regularization weight zeta. Over 500 paired Monte Carlo trials, forward KL is nearly insensitive to zeta, whereas reverse and symmetric objectives exhibit low-error regions followed by degradation under stronger regularization. At zeta = 3 x 10^-5, Average Jeffreys reduces median channel NRMSE by approximately 4.8% relative to forward KL at the same weight. Reduced KL asymmetry does not necessarily imply lower channel error. A separate fully blind QPSK receiver estimates a direct equalizer from received data only. Over 3000 paired trials, the VISTA variants have nearly identical median aligned block-NMSE, while their lower mean aligned block-NMSE relative to the CMA-plus-separation baseline is concentrated in difficult, large-error trials. Direct-W optimization with Adam reproduces most of this mean reduction. In the fully blind benchmark, the tested weak regularizers provide no resolved improvement over unregularized VISTA.

eess.SP↗

Quaternion-Valued Adaptive Projection Statistics for Neural Receivers in Dual-Polarization Free-Space Quantum Communications

Dual-polarization free-space quantum communication receivers produce two coupled complex heterodyne streams. We develop robust quaternion receiver methods for such links under fading, polarization mixing, calibration mismatch, impulsive contamination, and other non-Gaussian disturbances. The four polarization and quadrature components are encoded as quaternion-valued samples. We formulate an identifiable effective channel model, Quaternion-Valued Projection Statistics (QVPS), and Adaptive QVPS (AQVPS), which augments samplewise projection outlyingness with robust local second-order descriptors. Robust complex and quaternion channel estimators use radial Huber weighting, while QVPS and AQVPS are evaluated separately as disturbance diagnostics. Neural experiments compare matched real-, complex-, quaternion-, and widely linear complex-valued receivers. Robust estimation reduces sensitivity to non-Gaussian contamination. Local second-order descriptors improve detection of the tested localized disturbances relative to QVPS, and a descriptor-only ablation slightly outperforms full AQVPS. Complex LMMSE gives the lowest evaluation MSE in the clean Gaussian SNR sweep, whereas the Quaternion NN, among the compared receivers with their nominal activations, gives the lowest mean evaluation MSE in the tested contaminated sweeps. Activation-matched experiments indicate that much of the nominal Quaternion NN robustness is attributable to bounded activation rather than quaternion parameterization alone. The methods are evaluated at the receiver processing level and do not constitute a composable CV-QKD security proof.

eess.SP↗

Compressed LLM Reprogramming for Vision-Aided Beam Prediction in Vehicular Networks

Large Language Model (LLM) reprogramming-based beam prediction demonstrates strong data efficiency by adapting pretrained language models for vehicle-to-infrastructure (V2I) beam prediction, yet the resulting model complexity makes such approaches impractical for latency-sensitive deployment. We propose LLMBP-Lite, a compact LLM-reprogrammed framework for beam prediction. It leverages structural redundancy through pruning along three complementary dimensions: Transformer depth, source-prototype vocabulary size, and prompt length. Additionally, knowledge distillation can be optionally employed to maintain the pretrained representational capacity after compression. Experiments on the real-world dataset demonstrate that LLMBP-Lite achieves a $65\times$ inference speedup over the uncompressed LLM-based baseline while maintaining prediction accuracy and consistently outperforming recurrent baselines under limited training data. These results demonstrate that the tradeoff between data efficiency and deployment efficiency can be substantially mitigated in vehicular networks.

eess.SP↗

Budget-Constrained Graph Augmentation for Robust Network Design via Kirchhoff Index Minimization

Enhancing the robustness of deployed networks against failures and disruptions is critical for reliable operation. This requires deciding which new links to install and how strongly to weight them under limited resources. We study this problem through the Kirchhoff index, or total effective resistance, a spectral measure of global connectivity. The resulting augmentation problem couples discrete candidate-edge selection with continuous weight allocation under heterogeneous per-unit deployment costs, a total budget, and an exact-cardinality constraint. For a fixed weighted base graph, this yields a mixed-integer formulation and a semidefinite relaxation whose optimum lower-bounds the mixed-integer optimum. We cast the relaxation as a cone program and solve it numerically using a homogeneous self-dual embedding and first-order operator splitting. Feasible discrete designs are recovered through rounding-and-repair procedures and assessed by \emph{a posteriori} gap estimates relative to the numerical semidefinite program (SDP) benchmark. As a scalable alternative, we develop an exact-$k$, budget-feasible greedy heuristic built on rank-one Laplacian updates and biharmonic-distance caching, and interpret its progress through a Bellman value-to-go benchmark with a conservative spectral lower bound on the local policy ratio. Experiments on synthetic and real infrastructure networks across graph sizes, budgets, weight distributions, and cost regimes show that, under fixed budgets, distance-proportional costs limit the achievable resistance reduction and shift installed conductance toward shorter links relative to uniform per-unit costs.

math.OC↗

Communication-Efficient Byzantine-Robust Federated Conformal Prediction via Partial Sharing

We propose PRISM-FCP (Partial shaRing and robust calIbration with Statistical Margins for Federated Conformal Prediction), a communication-efficient Byzantine-robust federated conformal prediction framework that uses partial model sharing to mitigate stochastic model-poisoning attacks during training and histogram-based filtering to mitigate adversarial calibration submissions. Existing robust FCP approaches primarily address adversarial behavior during calibration, leaving training-stage poisoning to separate robust-training mechanisms. PRISM-FCP instead considers the coupling between the two stages. During training, clients partially share updates by transmitting only $M$ of $D$ parameters per round. Under the stated stochastic attack model, this attenuates the expected energy of each Byzantine client's perturbation contribution to the aggregate by a factor of $M/D$ relative to full sharing. When this benefit outweighs the optimization slowdown caused by partial updates, it can reduce training error and improve interval efficiency. During calibration, the server uses client-provided characterization vectors to filter suspected Byzantine clients before estimating the conformal quantile from the retained clients. Experiments on synthetic benchmarks and the UCI Superconductivity and YearPredictionMSD datasets, including Gaussian, ALIE, and sign-flipping training-stage attacks, demonstrate near-nominal empirical coverage and favorable communication--performance tradeoffs in the studied settings.

cs.LG↗

AdGT: Decentralized Gradient Tracking with Adaptive Per-Agent Stepsizes

In decentralized optimization, gradient-tracking methods typically rely on a single global stepsize. This choice can be conservative when agents have local objectives with different smoothness constants, since the stepsize must remain stable for the agent with the largest smoothness constant. This paper proposes AdGT, a decentralized gradient-tracking method in which each agent adapts its own stepsize using local gradient variation and a single global safety factor. The method reduces fixed-stepsize tuning effort and allows agents to exploit local smoothness information during the iterations. For smooth and strongly convex local objectives over undirected networks, we prove that the analyzed AdGT update converges linearly to the exact consensus optimizer. We also study two adaptive stepsize updates that use changes in the gradient-tracking direction. We characterize when the corresponding candidate determines the stepsize and prove conditional lower and upper stepsize bounds and linear convergence under an additional relative tracking-disagreement condition. Experiments on logistic regression, ridge regression, synthetic quadratic problems, and a linear-regression benchmark against state-of-the-art decentralized solvers show that AdGT often reaches a given accuracy in fewer iterations or gradient evaluations than tuned fixed-stepsize GT and the tested baselines, especially under heterogeneous local smoothness. In the topology experiments, each tested AdGT update uses one common safety factor across all graphs, whereas the fixed GT stepsize is tuned separately for each graph and seed.

math.OC↗

Partial Model Sharing Improves Byzantine Resilience in Federated Conformal Prediction

We propose a Byzantine-resilient federated conformal prediction (FCP) method that leverages partial model sharing, where only a subset of model parameters is exchanged each round. Unlike existing robust FCP approaches that primarily harden the calibration stage, our method protects both the federated training and conformal calibration phases. During training, partial sharing inherently restricts the attack surface and attenuates poisoned updates while reducing communication. During calibration, clients compress their non-conformity scores into histogram-based characterization vectors, enabling the server to detect Byzantine clients via distance-based maliciousness scores and to estimate the conformal quantile using only benign contributors. Experiments across diverse Byzantine attack scenarios show that the proposed method achieves closer-to-nominal coverage with substantially tighter prediction intervals than standard FCP, establishing a robust and communication-efficient approach to federated uncertainty quantification.

cs.LG↗

SA-PEF: Step-Ahead Partial Error Feedback for Efficient Federated Learning

Biased gradient compression with error feedback (EF) reduces communication in federated learning (FL), but under non-IID data, the residual error can decay slowly, causing gradient mismatch and stalled progress in the early rounds. We propose step-ahead partial error feedback (SA-PEF), which integrates step-ahead (SA) correction with partial error feedback (PEF). SA-PEF recovers EF when the step-ahead coefficient $α=0$ and step-ahead EF (SAEF) when $α=1$. For non-convex objectives and $δ$-contractive compressors, we establish a second-moment bound and a residual recursion that guarantee convergence to stationarity under heterogeneous data and partial client participation. The resulting rates match standard non-convex Fed-SGD guarantees up to constant factors, achieving $O((η,η_0TR)^{-1})$ convergence to a variance/heterogeneity floor with a fixed inner step size. Our analysis reveals a step-ahead-controlled residual contraction $ρ_r$ that explains the observed acceleration in the early training phase. To balance SAEF's rapid warm-up with EF's long-term stability, we select $α$ near its theory-predicted optimum. Experiments across diverse architectures and datasets show that SA-PEF consistently reaches target accuracy faster than EF.

cs.LG↗

Federated Smoothing ADMM for Localization

This paper addresses the challenge of localization in federated settings, which are characterized by distributed data, non-convexity, and non-smoothness. To tackle the scalability and outlier issues inherent in such environments, we propose a robust algorithm that employs an $\ell_1$-norm formulation within a novel federated ADMM framework. This approach addresses the problem by integrating an iterative smooth approximation for the total variation consensus term and employing a Moreau envelope approximation for the convex function that appears in a subtracted form. This transformation ensures that the problem is smooth and weakly convex in each iteration, which results in enhanced computational efficiency and improved estimation accuracy. The proposed algorithm supports asynchronous updates and multiple client updates per iteration, which ensures its adaptability to real-world federated systems. To validate the reliability of the proposed algorithm, we show that the method converges to a stationary point, and numerical simulations highlight its superior performance in convergence speed and outlier resilience compared to existing state-of-the-art localization methods.

cs.LG↗

Smoothing ADMM for Non-convex and Non-smooth Hierarchical Federated Learning

This paper presents a hierarchical federated learning (FL) framework that extends the alternating direction method of multipliers (ADMM) with smoothing techniques, tailored for non-convex and non-smooth objectives. Unlike traditional hierarchical FL methods, our approach supports asynchronous updates and multiple updates per iteration, enhancing adaptability to heterogeneous data and system settings. Additionally, we introduce a flexible mechanism to leverage diverse regularization functions at each layer, allowing customization to the specific prior information within each cluster and accommodating (possibly) non-smooth penalty objectives. Depending on the learning goal, the framework supports both consensus and personalization: the total variation norm can be used to enforce consensus across layers, while non-convex penalties such as minimax concave penalty (MCP) or smoothly clipped absolute deviation (SCAD) enable personalized learning. Experimental results demonstrate the superior convergence rates and accuracy of our method compared to conventional approaches, underscoring its robustness and versatility for a wide range of FL scenarios.

cs.LG↗

Noise-Robust and Resource-Efficient ADMM-based Federated Learning

Federated learning (FL) leverages client-server communications to train global models on decentralized data. However, communication noise or errors can impair model accuracy. To address this problem, we propose a novel FL algorithm that enhances robustness against communication noise while also reducing communication load. We derive the proposed algorithm through solving the weighted least-squares (WLS) regression problem as an illustrative example. We first frame WLS regression as a distributed convex optimization problem over a federated network employing random scheduling for improved communication efficiency. We then apply the alternating direction method of multipliers (ADMM) to iteratively solve this problem. To counteract the detrimental effects of cumulative communication noise, we introduce a key modification by eliminating the dual variable and implementing a new local model update at each participating client. This subtle yet effective change results in using a single noisy global model update at each client instead of two, improving robustness against additive communication noise. Furthermore, we incorporate another modification enabling clients to continue local updates even when not selected by the server, leading to substantial performance improvements. Our theoretical analysis confirms the convergence of our algorithm in both mean and the mean-square senses, even when the server communicates with a random subset of clients over noisy links at each iteration. Numerical results validate the effectiveness of our proposed algorithm and corroborate our theoretical findings.

cs.LG↗

Resilience in Online Federated Learning: Mitigating Model-Poisoning Attacks via Partial Sharing

Federated learning (FL) allows training machine learning models on distributed data without compromising privacy. However, FL is vulnerable to model-poisoning attacks where malicious clients tamper with their local models to manipulate the global model. In this work, we investigate the resilience of the partial-sharing online FL (PSO-Fed) algorithm against such attacks. PSO-Fed reduces communication overhead by allowing clients to share only a fraction of their model updates with the server. We demonstrate that this partial sharing mechanism has the added advantage of enhancing PSO-Fed's robustness to model-poisoning attacks. Through theoretical analysis, we show that PSO-Fed maintains convergence even under Byzantine attacks, where malicious clients inject noise into their updates. Furthermore, we derive a formula for PSO-Fed's mean square error, considering factors like stepsize, attack probability, and the number of malicious clients. Interestingly, we find a non-trivial optimal stepsize that maximizes PSO-Fed's resistance to these attacks. Extensive numerical experiments confirm our theoretical findings and showcase PSO-Fed's superior performance against model-poisoning attacks compared to other leading FL algorithms.

cs.LG↗

Federated Smoothing Proximal Gradient for Quantile Regression with Non-Convex Penalties

Distributed sensors in the internet-of-things (IoT) generate vast amounts of sparse data. Analyzing this high-dimensional data and identifying relevant predictors pose substantial challenges, especially when data is preferred to remain on the device where it was collected for reasons such as data integrity, communication bandwidth, and privacy. This paper introduces a federated quantile regression algorithm to address these challenges. Quantile regression provides a more comprehensive view of the relationship between variables than mean regression models. However, traditional approaches face difficulties when dealing with nonconvex sparse penalties and the inherent non-smoothness of the loss function. For this purpose, we propose a federated smoothing proximal gradient (FSPG) algorithm that integrates a smoothing mechanism with the proximal gradient framework, thereby enhancing both precision and computational speed. This integration adeptly handles optimization over a network of devices, each holding local data samples, making it particularly effective in federated learning scenarios. The FSPG algorithm ensures steady progress and reliable convergence in each iteration by maintaining or reducing the value of the objective function. By leveraging nonconvex penalties, such as the minimax concave penalty (MCP) and smoothly clipped absolute deviation (SCAD), the proposed method can identify and preserve key predictors within sparse models. Comprehensive simulations validate the robust theoretical foundations of the proposed algorithm and demonstrate improved estimation precision and reliable convergence.

cs.LG↗

Decentralized Smoothing ADMM for Quantile Regression with Non-Convex Sparse Penalties

In the rapidly evolving internet-of-things (IoT) ecosystem, effective data analysis techniques are crucial for handling distributed data generated by sensors. Addressing the limitations of existing methods, such as the sub-gradient approach, which fails to distinguish between active and non-active coefficients effectively, this paper introduces the decentralized smoothing alternating direction method of multipliers (DSAD) for penalized quantile regression. Our method leverages non-convex sparse penalties like the minimax concave penalty (MCP) and smoothly clipped absolute deviation (SCAD), improving the identification and retention of significant predictors. DSAD incorporates a total variation norm within a smoothing ADMM framework, achieving consensus among distributed nodes and ensuring uniform model performance across disparate data sources. This approach overcomes traditional convergence challenges associated with non-convex penalties in decentralized settings. We present theoretical proofs and extensive simulation results to validate the effectiveness of the DSAD, demonstrating its superiority in achieving reliable convergence and enhancing estimation accuracy compared with prior methods.

cs.LG↗

Distributed Maximum Consensus over Noisy Links

We introduce a distributed algorithm, termed noise-robust distributed maximum consensus (RD-MC), for estimating the maximum value within a multi-agent network in the presence of noisy communication links. Our approach entails redefining the maximum consensus problem as a distributed optimization problem, allowing a solution using the alternating direction method of multipliers. Unlike existing algorithms that rely on multiple sets of noise-corrupted estimates, RD-MC employs a single set, enhancing both robustness and efficiency. To further mitigate the effects of link noise and improve robustness, we apply moving averaging to the local estimates. Through extensive simulations, we demonstrate that RD-MC is significantly more robust to communication link noise compared to existing maximum-consensus algorithms.

cs.DC↗