计算机应用研究2024，Vol.41Issue(7)：2134-2140,7.DOI:10.19734/j.issn.1001-3695.2023.11.0562
跨脉冲传播的深度脉冲神经网络训练方法
Deep spiking neural network training method across spiking propagation
摘要
Abstract
Backpropagation-based training methods for SNNs still face many problems and challenges,including that the spike firing process is non-differentiable and spike neurons have complex spatiotemporal dynamics processes.In addition,SNNs backpropagation training methods often do not consider the relationship of the error signal between adjacent spikes,greatly re-ducing the accuracy of the model.To this end,this paper proposed a cross-spike error backpropagation training method for deep spiking neural networks(CSBP),which divided the error backpropagation of neurons into two dependencies:the depend-ency of spike firing time with the postsynaptic membrane potential(DSFT)and the dependency between spike firing time(DBSFT).Among them,DSFT solved the problem of spike non-differentiability and DBSFT clarified the dependence between spikes,allowing error signals to propagate across spikes,improving biological rationality.In addition,this paper solved the problem of insufficient expressive ability in early spiking ResNet network architecture by modifying the structural order of the spike residual block.Experimental results show that the proposed method is significantly improved compared to the SOTA(state-of-the-art)training algorithms based on spike time.Under the same architecture,the improvement is 2.98％on the CIFAR10 dataset,and 2.26％on the DVS-CIFAR10 dataset.关键词
脉冲神经网络/脉冲时间依赖/误差反向传播/脉冲神经网络训练算法Key words
spiking neural networks(SNNs)/spike time dependency/error backpropagation/spiking neural network training algorithm分类
信息技术与安全科学引用本文复制引用
曾建新,陈云华,李炜奇,陈平华..跨脉冲传播的深度脉冲神经网络训练方法[J].计算机应用研究,2024,41(7):2134-2140,7.基金项目
国家社会科学基金资助项目(20BKG031) （20BKG031）
