2408.09108

Total: 1

#1 Temporal Reversal Regularization for Spiking Neural Networks: Hybrid Spatio-Temporal Invariance for Generalization [PDF] [Copy] [Kimi] [REL]

Authors: Lin Zuo, Yongqi Ding, Wenwei Luo, Mengmeng Jing, Kunshan Yang

Spiking neural networks (SNNs) have received widespread attention as an ultra-low power computing paradigm. Recent studies have shown that SNNs suffer from severe overfitting, which limits their generalization performance. In this paper, we propose a simple yet effective Temporal Reversal Regularization (TRR) to mitigate overfitting during training and facilitate generalization of SNNs. We exploit the inherent temporal properties of SNNs to perform input/feature temporal reversal perturbations, prompting the SNN to produce original-reversed consistent outputs and learn perturbation-invariant representations. To further enhance generalization, we utilize the lightweight ``star operation" (Hadamard product) for feature hybridization of original and temporally reversed spike firing rates, which expands the implicit dimensionality and acts as a spatio-temporal regularizer. We show theoretically that our method is able to tighten the upper bound of the generalization error, and extensive experiments on static/neuromorphic recognition as well as 3D point cloud classification tasks demonstrate its effectiveness, versatility, and adversarial robustness. In particular, our regularization significantly improves the recognition accuracy of low-latency SNN for neuromorphic objects, contributing to the real-world deployment of neuromorphic computational software-hardware integration.

Subjects: Artificial Intelligence , Computer Vision and Pattern Recognition

Publish: 2024-08-17 06:23:38 UTC