Saved in:
Bibliographic Details
Main Authors: Guo, Yufei, Zhang, Yuhan, Jie, Zhou, Liu, Xiaode, Tong, Xin, Chen, Yuanpei, Peng, Weihang, Ma, Zhe
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2506.07720
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913885431267328
author Guo, Yufei
Zhang, Yuhan
Jie, Zhou
Liu, Xiaode
Tong, Xin
Chen, Yuanpei
Peng, Weihang
Ma, Zhe
author_facet Guo, Yufei
Zhang, Yuhan
Jie, Zhou
Liu, Xiaode
Tong, Xin
Chen, Yuanpei
Peng, Weihang
Ma, Zhe
contents The Spiking Neural Network (SNN), a biologically inspired neural network infrastructure, has garnered significant attention recently. SNNs utilize binary spike activations for efficient information transmission, replacing multiplications with additions, thereby enhancing energy efficiency. However, binary spike activation maps often fail to capture sufficient data information, resulting in reduced accuracy. To address this challenge, we advocate reversing the bit of the weight and activation for SNNs, called \textbf{ReverB-SNN}, inspired by recent findings that highlight greater accuracy degradation from quantizing activations compared to weights. Specifically, our method employs real-valued spike activations alongside binary weights in SNNs. This preserves the event-driven and multiplication-free advantages of standard SNNs while enhancing the information capacity of activations. Additionally, we introduce a trainable factor within binary weights to adaptively learn suitable weight amplitudes during training, thereby increasing network capacity. To maintain efficiency akin to vanilla \textbf{ReverB-SNN}, our trainable binary weight SNNs are converted back to standard form using a re-parameterization technique during inference. Extensive experiments across various network architectures and datasets, both static and dynamic, demonstrate that our approach consistently outperforms state-of-the-art methods.
format Preprint
id arxiv_https___arxiv_org_abs_2506_07720
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle ReverB-SNN: Reversing Bit of the Weight and Activation for Spiking Neural Networks
Guo, Yufei
Zhang, Yuhan
Jie, Zhou
Liu, Xiaode
Tong, Xin
Chen, Yuanpei
Peng, Weihang
Ma, Zhe
Computer Vision and Pattern Recognition
The Spiking Neural Network (SNN), a biologically inspired neural network infrastructure, has garnered significant attention recently. SNNs utilize binary spike activations for efficient information transmission, replacing multiplications with additions, thereby enhancing energy efficiency. However, binary spike activation maps often fail to capture sufficient data information, resulting in reduced accuracy. To address this challenge, we advocate reversing the bit of the weight and activation for SNNs, called \textbf{ReverB-SNN}, inspired by recent findings that highlight greater accuracy degradation from quantizing activations compared to weights. Specifically, our method employs real-valued spike activations alongside binary weights in SNNs. This preserves the event-driven and multiplication-free advantages of standard SNNs while enhancing the information capacity of activations. Additionally, we introduce a trainable factor within binary weights to adaptively learn suitable weight amplitudes during training, thereby increasing network capacity. To maintain efficiency akin to vanilla \textbf{ReverB-SNN}, our trainable binary weight SNNs are converted back to standard form using a re-parameterization technique during inference. Extensive experiments across various network architectures and datasets, both static and dynamic, demonstrate that our approach consistently outperforms state-of-the-art methods.
title ReverB-SNN: Reversing Bit of the Weight and Activation for Spiking Neural Networks
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2506.07720