On the Privacy Risks of Spiking Neural Networks: A Membership Inference Analysis

TL;DR

This paper investigates the privacy vulnerabilities of Spiking Neural Networks (SNNs) against Membership Inference Attacks, revealing that their robustness diminishes with increased latency and can be compromised using input dropout strategies.

Contribution

It provides the first comprehensive analysis of membership inference risks in SNNs and introduces an input dropout method to enhance attack success under black box settings.

Findings

SNNs are vulnerable to MIAs, especially at higher latency levels.

Input dropout significantly improves membership inference success.

SNN privacy risks are comparable to those of traditional ANNs.

Abstract

Spiking Neural Networks (SNNs) are increasingly explored for their energy efficiency and robustness in real-world applications, yet their privacy risks remain largely unexamined. In this work, we investigate the susceptibility of SNNs to Membership Inference Attacks (MIAs) -- a major privacy threat where an adversary attempts to determine whether a given sample was part of the training dataset. While prior work suggests that SNNs may offer inherent robustness due to their discrete, event-driven nature, we find that its resilience diminishes as latency (T) increases. Furthermore, we introduce an input dropout strategy under black box setting, that significantly enhances membership inference in SNNs. Our findings challenge the assumption that SNNs are inherently more secure, and even though they are expected to be better, our results reveal that SNNs exhibit privacy vulnerabilities that are equally comparable to Artificial Neural Networks (ANNs). Our code is available at https://github.com/sharmaabhijith/MIA_SNN.