Search for: All records

Risk-averse modeling is critical in safety-sensitive and high-stakes applications. Conditional Value-at-Risk (CVaR) quantifies such risk by measuring the expected loss in the tail of the loss distribution, and minimizing it provides a principled framework for training robust models. However, direct CVaR minimization re- mains challenging due to the difficulty of accurately estimating rare, high-loss events—particularly at extreme quantiles. In this work, we propose a novel train- ing framework that synthesizes informative samples for CVaR optimization using score-based generative models. Specifically, we guide a diffusion-based generative model to sample from a reweighted distribution that emphasizes inputs likely to incur high loss under a pretrained reference model. These samples are then in- corporated via a loss-weighted importance sampling scheme to reduce noise in stochastic optimization. We establish convergence guarantees and show that the synthesized, high-loss-emphasized dataset substantially contributes to the noise reduction. Empirically, we validate the effectiveness of our approach across mul- tiple settings, including a real-world wireless channel compression task, where our method achieves significant improvements over standard risk minimization strategies. 

Neural network-based encoders and decoders have demonstrated significant performance gains over traditional methods for Channel State Information (CSI) feedback in MIMO communications. However, key challenges in deploying these models in real-world scenarios remain underexplored, including: a) the need to efficiently accommodate diverse channel conditions across varying contexts, e.g., environments, and whether to use multiple encoders and decoders; b) the cost of gathering sufficient data to train neural network models across various contexts; and c) the need to protect sensitive data regarding competing providers’ coverages. To address the first challenge, we propose a novel system using context-dependent decoders and a universal encoder. We limit the number of decoders by clustering similar contexts and allowing those within a cluster to share the same decoder. To address the second and third challenges, we introduce a clustered federated learning-based approach that jointly clusters contexts and learns the desired encoder and context cluster-dependent decoders, leveraging distributed data. The clustering is performed efficiently based on the similarity of time-averaged gradients across contexts. To evaluate our approach, a new dataset reflecting the heterogeneous nature of the wireless systems was curated and made publicly available. Extensive experimental results demonstrate that our proposed CSI compression framework is highly effective and able to efficiently determine a correct context clustering and associated encoder and decoders. 

Progress in designing channel codes has been driven by human ingenuity and, fittingly, has been sporadic. Polar codes, developed on the foundation of Arikan's polarization kernel, represent the latest breakthrough in coding theory and have emerged as the state-of-the-art error-correction code for short-to-medium block length regimes. In an effort to automate the invention of good channel codes, especially in this regime, we explore a novel, non-linear generalization of Polar codes, which we call DEEPPOLAR codes. DEEPPOLAR codes extend the conventional Polar coding framework by utilizing a larger kernel size and parameterizing these kernels and matched decoders through neural networks. Our results demonstrate that these data-driven codes effectively leverage the benefits of a larger kernel size, resulting in enhanced reliability when compared to both existing neural codes and conventional Polar codes. 

In modern wireless systems, the feedback of DownLink (DL) Channel State Information (CSI) from User Equipment (UE) to Base Stations (BS) may require substantial computational and feedback bandwidth overheads. A promising approach to improve feedback efficiency is to leverage side information which is correlated to DL CSI. Despite potential of doing so, critical aspects remain underexplored in current research, particularly the quantification of the benefits and the inherent limitations of utilizing side information. This paper addresses these gaps by introducing a novel algorithm to compute the rate-distortion function for general compression scenarios incorporating side information. We apply this algorithm to the DL CSI feedback problem having UL CSI as the side information and generate rate-distortion functions. Using the estimated rate- distortion functions, we measure the gain of side information over diverse feedback rates and UE mobility profiles. The results reveal that the benefits of leveraging side information are particularly significant for UEs characterized by high mobility and constrained to operate at low feedback overheads. 

While neural lossy compression techniques have markedly advanced the efficiency of Channel State Information (CSI) compression and reconstruction for feedback in MIMO communications, efficient algorithms for more challenging and practical tasks—such as CSI compression for future channel pre- diction and reconstruction with relevant side information—remain underexplored, often resulting in suboptimal performance when existing methods are extended to these scenarios. To that end, we propose a novel framework for compression with side information, featuring an encoding process with fixed-rate compression using a trainable codebook for codeword quantization, and a decoding procedure modeled as a backward diffusion process conditioned on both the codeword and the side information. Experimental results show that our method significantly outperforms existing CSI compression algorithms, often yielding over twofold performance improvement by achieving comparable distortion at less than half the data rate of competing methods in certain scenarios. These findings underscore the potential of diffusion-based co