Quantized Spike-driven Transformer

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Qiu, Xuerui, Zhang, Malu, Zhang, Jieyuan, Wei, Wenjie, Cao, Honglin, Guo, Junsheng, Zhu, Rui-Jie, Shan, Yimeng, Yang, Yang, Li, Haizhou
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908279162011648
author Qiu, Xuerui
Zhang, Malu
Zhang, Jieyuan
Wei, Wenjie
Cao, Honglin
Guo, Junsheng
Zhu, Rui-Jie
Shan, Yimeng
Yang, Yang
Li, Haizhou
author_facet Qiu, Xuerui
Zhang, Malu
Zhang, Jieyuan
Wei, Wenjie
Cao, Honglin
Guo, Junsheng
Zhu, Rui-Jie
Shan, Yimeng
Yang, Yang
Li, Haizhou
contents Spiking neural networks are emerging as a promising energy-efficient alternative to traditional artificial neural networks due to their spike-driven paradigm. However, recent research in the SNN domain has mainly focused on enhancing accuracy by designing large-scale Transformer structures, which typically rely on substantial computational resources, limiting their deployment on resource-constrained devices. To overcome this challenge, we propose a quantized spike-driven Transformer baseline (QSD-Transformer), which achieves reduced resource demands by utilizing a low bit-width parameter. Regrettably, the QSD-Transformer often suffers from severe performance degradation. In this paper, we first conduct empirical analysis and find that the bimodal distribution of quantized spike-driven self-attention (Q-SDSA) leads to spike information distortion (SID) during quantization, causing significant performance degradation. To mitigate this issue, we take inspiration from mutual information entropy and propose a bi-level optimization strategy to rectify the information distribution in Q-SDSA. Specifically, at the lower level, we introduce an information-enhanced LIF to rectify the information distribution in Q-SDSA. At the upper level, we propose a fine-grained distillation scheme for the QSD-Transformer to align the distribution in Q-SDSA with that in the counterpart ANN. By integrating the bi-level optimization strategy, the QSD-Transformer can attain enhanced energy efficiency without sacrificing its high-performance advantage. For instance, when compared to the prior SNN benchmark on ImageNet, the QSD-Transformer achieves 80.3% top-1 accuracy, accompanied by significant reductions of 6.0$\times$ and 8.1$\times$ in power consumption and model size, respectively. Code is available at https://github.com/bollossom/QSD-Transformer.
format Preprint
id arxiv_https___arxiv_org_abs_2501_13492
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Quantized Spike-driven Transformer
Qiu, Xuerui
Zhang, Malu
Zhang, Jieyuan
Wei, Wenjie
Cao, Honglin
Guo, Junsheng
Zhu, Rui-Jie
Shan, Yimeng
Yang, Yang
Li, Haizhou
Computer Vision and Pattern Recognition
Spiking neural networks are emerging as a promising energy-efficient alternative to traditional artificial neural networks due to their spike-driven paradigm. However, recent research in the SNN domain has mainly focused on enhancing accuracy by designing large-scale Transformer structures, which typically rely on substantial computational resources, limiting their deployment on resource-constrained devices. To overcome this challenge, we propose a quantized spike-driven Transformer baseline (QSD-Transformer), which achieves reduced resource demands by utilizing a low bit-width parameter. Regrettably, the QSD-Transformer often suffers from severe performance degradation. In this paper, we first conduct empirical analysis and find that the bimodal distribution of quantized spike-driven self-attention (Q-SDSA) leads to spike information distortion (SID) during quantization, causing significant performance degradation. To mitigate this issue, we take inspiration from mutual information entropy and propose a bi-level optimization strategy to rectify the information distribution in Q-SDSA. Specifically, at the lower level, we introduce an information-enhanced LIF to rectify the information distribution in Q-SDSA. At the upper level, we propose a fine-grained distillation scheme for the QSD-Transformer to align the distribution in Q-SDSA with that in the counterpart ANN. By integrating the bi-level optimization strategy, the QSD-Transformer can attain enhanced energy efficiency without sacrificing its high-performance advantage. For instance, when compared to the prior SNN benchmark on ImageNet, the QSD-Transformer achieves 80.3% top-1 accuracy, accompanied by significant reductions of 6.0$\times$ and 8.1$\times$ in power consumption and model size, respectively. Code is available at https://github.com/bollossom/QSD-Transformer.
title Quantized Spike-driven Transformer
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2501.13492