Improving Spiking Sparse Recovery via Non-Convex Penalties

TL;DR

This paper introduces an adaptive spiking neural network algorithm that employs non-convex penalties for sparse recovery, significantly improving accuracy over traditional convex penalty methods.

Contribution

It proposes a novel adaptive algorithm for non-convex regularized optimization in spiking neural networks, with proven global convergence.

Findings

Enhanced recovery accuracy with adaptive non-convex penalties

Effective convergence analysis of the proposed method

Experimental validation shows significant improvements

Abstract

Compared with digital methods, sparse recovery based on spiking neural networks has great advantages like high computational efficiency and low power-consumption. However, current spiking algorithms cannot guarantee more accurate estimates since they are usually designed to solve the classical optimization with convex penalties, especially the $\ell_{1}$-norm. In fact, convex penalties are observed to underestimate the true solution in practice, while non-convex ones can avoid the underestimation. Inspired by this, we propose an adaptive version of spiking sparse recovery algorithm to solve the non-convex regularized optimization, and provide an analysis on its global asymptotic convergence. Through experiments, the accuracy is greatly improved under different adaptive ways.