Unveiling the Power of Sparse Neural Networks for Feature Selection

TL;DR

This paper systematically analyzes sparse neural networks for feature selection, introduces a new metric for importance, and demonstrates significant efficiency gains and improved feature quality over dense networks.

Contribution

It provides a comprehensive analysis of DST algorithms in SNNs, introduces a novel feature importance metric, and compares performance across datasets.

Findings

Over 50% memory reduction compared to dense networks

More than 55% FLOPs reduction

Outperforms dense networks in feature quality

Abstract

Sparse Neural Networks (SNNs) have emerged as powerful tools for efficient feature selection. Leveraging the dynamic sparse training (DST) algorithms within SNNs has demonstrated promising feature selection capabilities while drastically reducing computational overheads. Despite these advancements, several critical aspects remain insufficiently explored for feature selection. Questions persist regarding the choice of the DST algorithm for network training, the choice of metric for ranking features/neurons, and the comparative performance of these methods across diverse datasets when compared to dense networks. This paper addresses these gaps by presenting a comprehensive systematic analysis of feature selection with sparse neural networks. Moreover, we introduce a novel metric considering sparse neural network characteristics, which is designed to quantify feature importance within the context of SNNs. Our findings show that feature selection with SNNs trained with DST algorithms can achieve, on average, more than $50\%$ memory and $55\%$ FLOPs reduction compared to the dense networks, while outperforming them in terms of the quality of the selected features. Our code and the supplementary material are available on GitHub (\url{https://github.com/zahraatashgahi/Neuron-Attribution}).