arXiv subjects

Swatantra Kafle

Publications and source records attributed to Swatantra Kafle.

5 recordsLinked to original sources

arXiv2025

One-bit Compressed Sensing using Generative Models

This paper addresses the classical problem of one-bit compressed sensing using a deep learning-based reconstruction algorithm that leverages a trained generative model to enhance the signal reconstruction performance. The generator, a pre-trained neural network, learns to map from a low-dimensional latent space to a higher-dimensional set of sparse vectors. This generator is then used to reconstruct sparse vectors from their one-bit measurements by searching over its range. The presented algorithm provides an excellent reconstruction performance because the generative model can learn additional structural information about the signal beyond sparsity. Furthermore, we provide theoretical guarantees on the reconstruction accuracy and sample complexity of the algorithm. Through numerical experiments using three publicly available image datasets, MNIST, Fashion-MNIST, and Omniglot, we demonstrate the superior performance of the algorithm compared to other existing algorithms and show that our algorithm can recover both the amplitude and the direction of the signal from one-bit measurements.

cs.LG↗

arXiv2023

Generalization of Deep Reinforcement Learning for Jammer-Resilient Frequency and Power Allocation

We tackle the problem of joint frequency and power allocation while emphasizing the generalization capability of a deep reinforcement learning model. Most of the existing methods solve reinforcement learning-based wireless problems for a specific pre-determined wireless network scenario. The performance of a trained agent tends to be very specific to the network and deteriorates when used in a different network operating scenario (e.g., different in size, neighborhood, and mobility, among others). We demonstrate our approach to enhance training to enable a higher generalization capability during inference of the deployed model in a distributed multi-agent setting in a hostile jamming environment. With all these, we show the improved training and inference performance of the proposed methods when tested on previously unseen simulated wireless networks of different sizes and architectures. More importantly, to prove practical impact, the end-to-end solution was implemented on the embedded software-defined radio and validated using over-the-air evaluation.

cs.NI↗

arXiv2020

Parallel Restarted SPIDER -- Communication Efficient Distributed Nonconvex Optimization with Optimal Computation Complexity

In this paper, we propose a distributed algorithm for stochastic smooth, non-convex optimization. We assume a worker-server architecture where $N$ nodes, each having $n$ (potentially infinite) number of samples, collaborate with the help of a central server to perform the optimization task. The global objective is to minimize the average of local cost functions available at individual nodes. The proposed approach is a non-trivial extension of the popular parallel-restarted SGD algorithm, incorporating the optimal variance-reduction based SPIDER gradient estimator into it. We prove convergence of our algorithm to a first-order stationary solution. The proposed approach achieves the best known communication complexity $O(ε^{-1})$ along with the optimal computation complexity. For finite-sum problems (finite $n$), we achieve the optimal computation (IFO) complexity $O(\sqrt{Nn}ε^{-1})$. For online problems ($n$ unknown or infinite), we achieve the optimal IFO complexity $O(ε^{-3/2})$. In both the cases, we maintain the linear speedup achieved by existing methods. This is a massive improvement over the $O(ε^{-2})$ IFO complexity of the existing approaches. Additionally, our algorithm is general enough to allow non-identical distributions of data across workers, as in the recently proposed federated learning paradigm.

math.OC↗

arXiv2020

Noisy One-bit Compressed Sensing with Side-Information

We consider the problem of sparse signal reconstruction from noisy one-bit compressed measurements when the receiver has access to side-information (SI). We assume that compressed measurements are corrupted by additive white Gaussian noise before quantization and sign-flip error after quantization. A generalized approximate message passing-based method for signal reconstruction from noisy one-bit compressed measurements is proposed, which is then extended for the case where the receiver has access to a signal that aids signal reconstruction, i.e., side-information. Two different scenarios of side-information are considered-a) side-information consisting of support information only, and b) side information consisting of support and amplitude information. SI is either a noisy version of the signal or a noisy estimate of the support of the signal. We develop reconstruction algorithms from one-bit measurements using noisy SI available at the receiver. Laplacian distribution and Bernoulli distribution are used to model the two types of noises which, when applied to the signal and the support, yields the SI for the above two cases, respectively. The Expectation-Maximization algorithm is used to estimate the noise parameters using noisy one-bit compressed measurements and the SI. We show that one-bit compressed measurement-based signal reconstruction is quite sensitive to noise, and the reconstruction performance can be significantly improved by exploiting available side-information at the receiver.

eess.SP↗

arXiv2020

Distributed Stochastic Non-Convex Optimization: Momentum-Based Variance Reduction

In this work, we propose a distributed algorithm for stochastic non-convex optimization. We consider a worker-server architecture where a set of $K$ worker nodes (WNs) in collaboration with a server node (SN) jointly aim to minimize a global, potentially non-convex objective function. The objective function is assumed to be the sum of local objective functions available at each WN, with each node having access to only the stochastic samples of its local objective function. In contrast to the existing approaches, we employ a momentum based "single loop" distributed algorithm which eliminates the need of computing large batch size gradients to achieve variance reduction. We propose two algorithms one with "adaptive" and the other with "non-adaptive" learning rates. We show that the proposed algorithms achieve the optimal computational complexity while attaining linear speedup with the number of WNs. Specifically, the algorithms reach an $ε$-stationary point $x_a$ with $\mathbb{E}\| \nabla f(x_a) \| \leq \tilde{O}(K^{-1/3}T^{-1/2} + K^{-1/3}T^{-1/3})$ in $T$ iterations, thereby requiring $\tilde{O}(K^{-1} ε^{-3})$ gradient computations at each WN. Moreover, our approach does not assume identical data distributions across WNs making the approach general enough for federated learning applications.

math.OC↗