Watermarking Neuromorphic Brains: Intellectual Property Protection in Spiking Neural Networks

TL;DR

This paper explores watermarking techniques to protect intellectual property in spiking neural networks, addressing unique challenges and evaluating robustness against various attacks.

Contribution

It adapts and evaluates fingerprint-based and backdoor-based watermarking methods specifically for SNNs, a novel application area.

Findings

Watermarking methods show promise for SNN IP protection.

Resilience varies depending on attack type and method.

Comparison with ANN watermarking highlights unique SNN challenges.

Abstract

As spiking neural networks (SNNs) gain traction in deploying neuromorphic computing solutions, protecting their intellectual property (IP) has become crucial. Without adequate safeguards, proprietary SNN architectures are at risk of theft, replication, or misuse, which could lead to significant financial losses for the owners. While IP protection techniques have been extensively explored for artificial neural networks (ANNs), their applicability and effectiveness for the unique characteristics of SNNs remain largely unexplored. In this work, we pioneer an investigation into adapting two prominent watermarking approaches, namely, fingerprint-based and backdoor-based mechanisms to secure proprietary SNN architectures. We conduct thorough experiments to evaluate the impact on fidelity, resilience against overwrite threats, and resistance to compression attacks when applying these watermarking techniques to SNNs, drawing comparisons with their ANN counterparts. This study lays the groundwork for developing neuromorphic-aware IP protection strategies tailored to the distinctive dynamics of SNNs.