Are Neuromorphic Architectures Inherently Privacy-preserving? An Exploratory Study

2024-11-10

arXiv.org

AI Summary

This work investigates whether spiking neural networks (SNNs) intrinsically exhibit superior privacy robustness over artificial neural networks (ANNs) against membership inference attacks (MIAs). We conduct a systematic empirical evaluation across CIFAR-10 and CIFAR-100 using diverse SNN frameworks (snnTorch, TENNLab, LAVA), learning paradigms (surrogate gradient descent and evolutionary learning), and differentially private training (DPSGD). Our key findings are threefold: (i) SNNs achieve significantly lower MIA success—AUC scores of 0.59/0.58 versus 0.82/0.88 for ANNs—without any additional privacy mechanisms; (ii) evolutionary learning further enhances this inherent robustness; and (iii) under identical differential privacy constraints, SNNs incur substantially smaller utility degradation than ANNs. These results provide the first empirical evidence that neuromorphic computation possesses intrinsic privacy advantages, suggesting a promising low-overhead pathway toward secure AI in privacy-sensitive applications.

Technology Category

• Machine Learning: Privacy
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Computer Vision: Adversarial Attacks & Robustness

Application Category

• Security and Privacy: Security and privacy of machine learning and AI applications
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

While machine learning (ML) models are becoming mainstream, especially in sensitive application areas, the risk of data leakage has become a growing concern. Attacks like membership inference (MIA) have shown that trained models can reveal sensitive data, jeopardizing confidentiality. While traditional Artificial Neural Networks (ANNs) dominate ML applications, neuromorphic architectures, specifically Spiking Neural Networks (SNNs), are emerging as promising alternatives due to their low power consumption and event-driven processing, akin to biological neurons. Privacy in ANNs is well-studied; however, little work has explored the privacy-preserving properties of SNNs. This paper examines whether SNNs inherently offer better privacy. Using MIAs, we assess the privacy resilience of SNNs versus ANNs across diverse datasets. We analyze the impact of learning algorithms (surrogate gradient and evolutionary), frameworks (snnTorch, TENNLab, LAVA), and parameters on SNN privacy. Our findings show that SNNs consistently outperform ANNs in privacy preservation, with evolutionary algorithms offering additional resilience. For instance, on CIFAR-10, SNNs achieve an AUC of 0.59, significantly lower than ANNs' 0.82, and on CIFAR-100, SNNs maintain an AUC of 0.58 compared to ANNs' 0.88. Additionally, we explore the privacy-utility trade-off with Differentially Private Stochastic Gradient Descent (DPSGD), finding that SNNs sustain less accuracy loss than ANNs under similar privacy constraints.

Problem

• Explores privacy-preserving properties of Spiking Neural Networks.
• Compares privacy resilience between SNNs and traditional ANNs.
• Investigates impact of algorithms and frameworks on SNN privacy.

Innovation

• SNNs enhance privacy preservation
• Evolutionary algorithms boost SNN resilience
• SNNs outperform ANNs in privacy