Faster and Stronger: When ANN-SNN Conversion Meets Parallel Spiking Calculation

TL;DR

This paper introduces a novel parallel conversion framework for Spiking Neural Networks that reduces training overhead and inference latency, enabling efficient, scalable, and generalizable SNN deployment with theoretical validation and extensive experiments.

Contribution

It proposes a mathematically validated parallel conversion method combined with error calibration, pioneering the joint use of parallel spiking calculation and ANN-SNN conversion.

Findings

Achieves ultra-low latency conversion with significant performance gains.

Validates lossless and sorting properties of the conversion process.

Demonstrates effectiveness across various network configurations.

Abstract

Spiking Neural Network (SNN), as a brain-inspired and energy-efficient network, is currently facing the pivotal challenge of exploring a suitable and efficient learning framework. The predominant training methodologies, namely Spatial-Temporal Back-propagation (STBP) and ANN-SNN Conversion, are encumbered by substantial training overhead or pronounced inference latency, which impedes the advancement of SNNs in scaling to larger networks and navigating intricate application domains. In this work, we propose a novel parallel conversion learning framework, which establishes a mathematical mapping relationship between each time-step of the parallel spiking neurons and the cumulative spike firing rate. We theoretically validate the lossless and sorting properties of the conversion process, as well as pointing out the optimal shifting distance for each step. Furthermore, by integrating the above framework with the distribution-aware error calibration technique, we can achieve efficient conversion towards more general activation functions or training-free circumstance. Extensive experiments have confirmed the significant performance advantages of our method for various conversion cases under ultra-low time latency. To our best knowledge, this is the first work which jointly utilizes parallel spiking calculation and ANN-SNN Conversion, providing a highly promising approach for SNN supervised training. Code is available at https://github.com/hzc1208/Parallel_Conversion.