NeuroMoCo: A Neuromorphic Momentum Contrast Learning Method for Spiking Neural Networks

TL;DR

NeuroMoCo introduces a pioneering self-supervised momentum contrast learning method for spiking neural networks, significantly enhancing classification accuracy on neuromorphic datasets by leveraging a novel loss function and pre-training approach.

Contribution

This paper is the first to implement self-supervised learning based on momentum contrast in spiking neural networks, introducing a new loss function and achieving state-of-the-art results.

Findings

Achieved new SOTA benchmarks on multiple neuromorphic datasets.

Demonstrated effectiveness of MixInfoNCE loss in temporal feature extraction.

Validated the benefits of self-supervised pre-training for SNNs.

Abstract

Recently, brain-inspired spiking neural networks (SNNs) have attracted great research attention owing to their inherent bio-interpretability, event-triggered properties and powerful perception of spatiotemporal information, which is beneficial to handling event-based neuromorphic datasets. In contrast to conventional static image datasets, event-based neuromorphic datasets present heightened complexity in feature extraction due to their distinctive time series and sparsity characteristics, which influences their classification accuracy. To overcome this challenge, a novel approach termed Neuromorphic Momentum Contrast Learning (NeuroMoCo) for SNNs is introduced in this paper by extending the benefits of self-supervised pre-training to SNNs to effectively stimulate their potential. This is the first time that self-supervised learning (SSL) based on momentum contrastive learning is realized in SNNs. In addition, we devise a novel loss function named MixInfoNCE tailored to their temporal characteristics to further increase the classification accuracy of neuromorphic datasets, which is verified through rigorous ablation experiments. Finally, experiments on DVS-CIFAR10, DVS128Gesture and N-Caltech101 have shown that NeuroMoCo of this paper establishes new state-of-the-art (SOTA) benchmarks: 83.6% (Spikformer-2-256), 98.62% (Spikformer-2-256), and 84.4% (SEW-ResNet-18), respectively.