One-Timestep is Enough: Achieving High-performance ANN-to-SNN Conversion via Scale-and-Fire Neurons

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Chen, Qiuyang, Yang, Huiqi, Meng, Qingyan, Ma, Zhengyu
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866909871145746432
author Chen, Qiuyang
Yang, Huiqi
Meng, Qingyan
Ma, Zhengyu
author_facet Chen, Qiuyang
Yang, Huiqi
Meng, Qingyan
Ma, Zhengyu
contents Spiking Neural Networks (SNNs) are gaining attention as energy-efficient alternatives to Artificial Neural Networks (ANNs), especially in resource-constrained settings. While ANN-to-SNN conversion (ANN2SNN) achieves high accuracy without end-to-end SNN training, existing methods rely on large time steps, leading to high inference latency and computational cost. In this paper, we propose a theoretical and practical framework for single-timestep ANN2SNN. We establish the Temporal-to-Spatial Equivalence Theory, proving that multi-timestep integrate-and-fire (IF) neurons can be equivalently replaced by single-timestep multi-threshold neurons (MTN). Based on this theory, we introduce the Scale-and-Fire Neuron (SFN), which enables effective single-timestep ($T=1$) spiking through adaptive scaling and firing. Furthermore, we develop the SFN-based Spiking Transformer (SFormer), a specialized instantiation of SFN within Transformer architectures, where spike patterns are aligned with attention distributions to mitigate the computational, energy, and hardware overhead of the multi-threshold design. Extensive experiments on image classification, object detection, and instance segmentation demonstrate that our method achieves state-of-the-art performance under single-timestep inference. Notably, we achieve 88.8% top-1 accuracy on ImageNet-1K at $T=1$, surpassing existing conversion methods.
format Preprint
id arxiv_https___arxiv_org_abs_2510_23383
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle One-Timestep is Enough: Achieving High-performance ANN-to-SNN Conversion via Scale-and-Fire Neurons
Chen, Qiuyang
Yang, Huiqi
Meng, Qingyan
Ma, Zhengyu
Neural and Evolutionary Computing
Spiking Neural Networks (SNNs) are gaining attention as energy-efficient alternatives to Artificial Neural Networks (ANNs), especially in resource-constrained settings. While ANN-to-SNN conversion (ANN2SNN) achieves high accuracy without end-to-end SNN training, existing methods rely on large time steps, leading to high inference latency and computational cost. In this paper, we propose a theoretical and practical framework for single-timestep ANN2SNN. We establish the Temporal-to-Spatial Equivalence Theory, proving that multi-timestep integrate-and-fire (IF) neurons can be equivalently replaced by single-timestep multi-threshold neurons (MTN). Based on this theory, we introduce the Scale-and-Fire Neuron (SFN), which enables effective single-timestep ($T=1$) spiking through adaptive scaling and firing. Furthermore, we develop the SFN-based Spiking Transformer (SFormer), a specialized instantiation of SFN within Transformer architectures, where spike patterns are aligned with attention distributions to mitigate the computational, energy, and hardware overhead of the multi-threshold design. Extensive experiments on image classification, object detection, and instance segmentation demonstrate that our method achieves state-of-the-art performance under single-timestep inference. Notably, we achieve 88.8% top-1 accuracy on ImageNet-1K at $T=1$, surpassing existing conversion methods.
title One-Timestep is Enough: Achieving High-performance ANN-to-SNN Conversion via Scale-and-Fire Neurons
topic Neural and Evolutionary Computing
url https://arxiv.org/abs/2510.23383