Gradient Scaling on Deep Spiking Neural Networks with Spike-Dependent Local Information

TL;DR

This paper introduces a gradient scaling method based on local spike information to improve the training efficiency and accuracy of deep spiking neural networks, enabling better utilization of spike data.

Contribution

It proposes a novel gradient scaling technique that leverages spike causality, enhancing deep SNN training performance over existing methods.

Findings

Higher accuracy on CIFAR10 and CIFAR100 datasets.

Achieved lower spike counts while maintaining performance.

Enhanced training stability and efficiency.

Abstract

Deep spiking neural networks (SNNs) are promising neural networks for their model capacity from deep neural network architecture and energy efficiency from SNNs' operations. To train deep SNNs, recently, spatio-temporal backpropagation (STBP) with surrogate gradient was proposed. Although deep SNNs have been successfully trained with STBP, they cannot fully utilize spike information. In this work, we proposed gradient scaling with local spike information, which is the relation between pre- and post-synaptic spikes. Considering the causality between spikes, we could enhance the training performance of deep SNNs. According to our experiments, we could achieve higher accuracy with lower spikes by adopting the gradient scaling on image classification tasks, such as CIFAR10 and CIFAR100.