Spatio-Temporal Backpropagation for Training High-Performance Spiking Neural Networks

TLDR

Spiking neural networks encode richer spatio‑temporal information than artificial neural networks, yet current back‑propagation methods exploit only spatial cues and struggle with non‑differentiable spikes, limiting performance. The authors aim to develop an iterative leaky‑integrate‑and‑fire model that facilitates gradient‑descent training of SNNs. They propose a spatio‑temporal backpropagation framework that jointly optimizes spatial and temporal domains using an approximated spike derivative. The framework achieves state‑of‑the‑art performance on static MNIST, dynamic N‑MNIST, and a custom object‑detection dataset, demonstrating the potential of high‑performance SNNs for brain‑like computing.

Abstract

Compared with artificial neural networks (ANNs), spiking neural networks (SNNs) are promising to explore the brain-like behaviors since the spikes could encode more spatio-temporal information. Although pre-training from ANN or direct training based on backpropagation (BP) makes the supervised training of SNNs possible, these methods only exploit the networks' spatial domain information which leads to the performance bottleneck and requires many complicated training skills. Another fundamental issue is that the spike activity is naturally non-differentiable which causes great difficulties in training SNNs. To this end, we build an iterative LIF model that is more friendly for gradient descent training. By simultaneously considering the layer-by-layer spatial domain (SD) and the timing-dependent temporal domain (TD) in the training phase, as well as an approximated derivative for the spike activity, we propose a spatio-temporal backpropagation (STBP) training framework without using any complicated technology. We achieve the best performance of multi-layered perceptron (MLP) compared with existing state-of-the-art algorithms over the static MNIST and the dynamic N-MNIST dataset as well as a custom object detection dataset. This work provides a new perspective to explore the high-performance SNNs for future brain-like computing paradigm with rich spatio-temporal dynamics.