Ultra-low Latency Spiking Neural Networks with Spatio-Temporal Compression and Synaptic Convolutional Block

TL;DR

This paper introduces a novel spatio-temporal compression technique and synaptic convolutional block for spiking neural networks, significantly reducing latency while maintaining high accuracy in event stream classification.

Contribution

It proposes a spatio-temporal compression method combined with a synaptic convolutional block and multi-threshold LIF neurons to enhance SNN efficiency and accuracy.

Findings

Outperforms state-of-the-art accuracy on neuromorphic datasets

Reduces training and inference latency significantly

Achieves high accuracy with fewer time steps

Abstract

Spiking neural networks (SNNs), as one of the brain-inspired models, has spatio-temporal information processing capability, low power feature, and high biological plausibility. The effective spatio-temporal feature makes it suitable for event streams classification. However, neuromorphic datasets, such as N-MNIST, CIFAR10-DVS, DVS128-gesture, need to aggregate individual events into frames with a new higher temporal resolution for event stream classification, which causes high training and inference latency. In this work, we proposed a spatio-temporal compression method to aggregate individual events into a few time steps of synaptic current to reduce the training and inference latency. To keep the accuracy of SNNs under high compression ratios, we also proposed a synaptic convolutional block to balance the dramatic change between adjacent time steps. And multi-threshold Leaky Integrate-and-Fire (LIF) with learnable membrane time constant is introduced to increase its information processing capability. We evaluate the proposed method for event streams classification tasks on neuromorphic N-MNIST, CIFAR10-DVS, DVS128 gesture datasets. The experiment results show that our proposed method outperforms the state-of-the-art accuracy on nearly all datasets, using fewer time steps.