Temporal-adaptive Weight Quantization for Spiking Neural Networks

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Zhang, Han, Meng, Qingyan, Wang, Jiaqi, Chen, Baiyu, Ma, Zhengyu, Fan, Xiaopeng
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866917098087776256
author Zhang, Han
Meng, Qingyan
Wang, Jiaqi
Chen, Baiyu
Ma, Zhengyu
Fan, Xiaopeng
author_facet Zhang, Han
Meng, Qingyan
Wang, Jiaqi
Chen, Baiyu
Ma, Zhengyu
Fan, Xiaopeng
contents Weight quantization in spiking neural networks (SNNs) could further reduce energy consumption. However, quantizing weights without sacrificing accuracy remains challenging. In this study, inspired by astrocyte-mediated synaptic modulation in the biological nervous systems, we propose Temporal-adaptive Weight Quantization (TaWQ), which incorporates weight quantization with temporal dynamics to adaptively allocate ultra-low-bit weights along the temporal dimension. Extensive experiments on static (e.g., ImageNet) and neuromorphic (e.g., CIFAR10-DVS) datasets demonstrate that our TaWQ maintains high energy efficiency (4.12M, 0.63mJ) while incurring a negligible quantization loss of only 0.22% on ImageNet.
format Preprint
id arxiv_https___arxiv_org_abs_2511_17567
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Temporal-adaptive Weight Quantization for Spiking Neural Networks
Zhang, Han
Meng, Qingyan
Wang, Jiaqi
Chen, Baiyu
Ma, Zhengyu
Fan, Xiaopeng
Neural and Evolutionary Computing
Artificial Intelligence
Computer Vision and Pattern Recognition
Weight quantization in spiking neural networks (SNNs) could further reduce energy consumption. However, quantizing weights without sacrificing accuracy remains challenging. In this study, inspired by astrocyte-mediated synaptic modulation in the biological nervous systems, we propose Temporal-adaptive Weight Quantization (TaWQ), which incorporates weight quantization with temporal dynamics to adaptively allocate ultra-low-bit weights along the temporal dimension. Extensive experiments on static (e.g., ImageNet) and neuromorphic (e.g., CIFAR10-DVS) datasets demonstrate that our TaWQ maintains high energy efficiency (4.12M, 0.63mJ) while incurring a negligible quantization loss of only 0.22% on ImageNet.
title Temporal-adaptive Weight Quantization for Spiking Neural Networks
topic Neural and Evolutionary Computing
Artificial Intelligence
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2511.17567