SDSNN: A Single-Timestep Spiking Neural Network with Self-Dropping Neuron and Bayesian Optimization

TL;DR

This paper introduces a single-timestep spiking neural network with a self-dropping neuron mechanism and Bayesian optimization, achieving high accuracy and energy efficiency on standard datasets, thus enabling faster and more energy-efficient edge computing applications.

Contribution

It proposes a novel single-timestep SNN with dynamic neuron thresholds and Bayesian parameter optimization, reducing inference latency and energy consumption while maintaining high accuracy.

Findings

Achieves over 93% accuracy on Fashion-MNIST with a single timestep.

Reduces energy consumption by up to 56%.

Outperforms traditional multi-timestep SNNs in accuracy and efficiency.

Abstract

Spiking Neural Networks (SNNs), as an emerging biologically inspired computational model, demonstrate significant energy efficiency advantages due to their event-driven information processing mechanism. Compared to traditional Artificial Neural Networks (ANNs), SNNs transmit information through discrete spike signals, which substantially reduces computational energy consumption through their sparse encoding approach. However, the multi-timestep computation model significantly increases inference latency and energy, limiting the applicability of SNNs in edge computing scenarios. We propose a single-timestep SNN, which enhances accuracy and reduces computational energy consumption in a single timestep by optimizing spike generation and temporal parameters. We design a Self-Dropping Neuron mechanism, which enhances information-carrying capacity through dynamic threshold adjustment and selective spike suppression. Furthermore, we employ Bayesian optimization to globally search for time parameters and obtain an efficient inference mode with a single time step. Experimental results on the Fashion-MNIST, CIFAR-10, and CIFAR-100 datasets demonstrate that, compared to traditional multi-timestep SNNs employing the Leaky Integrate-and-Fire (LIF) model, our method achieves classification accuracies of 93.72%, 92.20%, and 69.45%, respectively, using only single-timestep spikes, while maintaining comparable or even superior accuracy. Additionally, it reduces energy consumption by 56%, 21%, and 22%, respectively.