一种脉冲神经元监督学习的直接计算方法

doi:10.12263/DZXB.20190350

电子学报 ›› 2021, Vol. 49 ›› Issue (2): 331-337.DOI: 10.12263/DZXB.20190350

一种脉冲神经元监督学习的直接计算方法

陈国军¹, 蔺想红², 王国恩¹, 王向文²

1. 武汉大学城市设计学院, 湖北武汉 430072;
2. 西北师范大学计算机科学与工程学院, 甘肃兰州 730070

收稿日期:2019-04-02 修回日期:2020-04-27 出版日期:2021-02-25
- 通讯作者:
- 王国恩
- 作者简介:
- 陈国军 男,1991年7月出生,山东德州人.2016年毕业于西北师范大学计算机科学与工程学院,获软件工程硕士学位.现为武汉大学城市设计学院博士生.研究方向为神经网络、智慧城市与城市规划.E-mail:guojunchen@whu.edu.cn;蔺想红 男,1976年1月出生,甘肃天水人.2009年获哈尔滨工业大学计算机应用技术专业博士学位.现任西北师范大学计算机科学与工程学院教授,硕士生导师.研究方向为神经网络、深度学习、智能软件.E-mail:linxh@nwnu.edu.cn;王向文 男,1991年3月出生,甘肃天水人.2015年毕业于西北师范大学,获得硕士学位.现任西北师范大学计算机科学与工程学院教师.研究方向为神经网络、机器学习与生物信息学.E-mail:wangxiangwen2@163.com
- 基金资助:
- 国家自然科学基金 (No.51878516）

A Direct Computation Method of Supervised Learning for Spiking Neurons

CHEN Guo-jun¹, LIN Xiang-hong², WANG Guo-en¹, WANG Xiang-wen²

1. School of Urban Design, Wuhan University, Wuhan, Hubei 430072, China;
2. Computer Science and Engineering, Northwest Normal University, Lanzhou, Gansu 730070, China

Received:2019-04-02 Revised:2020-04-27 Online:2021-02-25 Published:2021-02-25

摘要/Abstract

摘要： 精确脉冲定时作为一种神经元信息编码方式更具生物可解释性，使用精确脉冲定时编码的脉冲神经元具有更为强大的时空信号处理能力.脉冲神经元监督学习是神经计算的重要方面，目的是使神经元对给定输入脉冲在期望时刻发放脉冲.通过分析输入脉冲序列、期望输出脉冲序列与实际输出脉冲序列的关系，发现已有脉冲神经元监督学习算法的脉冲选择与计算较为复杂，致使不能达到理想学习效果.通过去除影响整体学习效果的多余脉冲计算，构建用于脉冲神经元突触权值调整的双脉冲单元，提出了一种适用于脉冲神经元监督学习的直接计算方法.该方法基于输入脉冲，使用期望输出脉冲与实际输出脉冲的时序关系，直接计算突触权值的调整量；每个输入脉冲在每次迭代中最多计算一次，有效减少了脉冲计算次数.实验结果表明，直接计算方法作为脉冲神经元监督学习的一般性脉冲计算优化策略，可以大幅提高已有算法的学习准确率.

关键词: 监督学习, 脉冲神经元, 脉冲神经网络, 脉冲序列, 直接计算

Abstract: The purpose of supervised learning for spiking neurons is to emit spikes at desired times.By analyzing the relationship between desired output spike train,actual output spike train,and input spike trains,we argue that the spike selection and calculation of existing algorithms are much complex.By removing the repetitive spike computations which have a negative effect on the overall learning performance,we construct a unit of pair-spike and propose a direct computation method of supervised learning algorithms for spiking neurons.Based on input spikes and the relationship between desired output spike train and actual output spike train,the proposed method utilities every useful input spike selected only once to directly adjust the synaptic weights,which truly reduces the computational cost.Experimental results show that the direct computation method,as a general optimization strategy of supervised learning for spiking neurons,can effectively improve the learning performance of existing algorithms.

Key words: supervised learning, spiking neuron, spiking neural networks, spike train, direct computation

中图分类号:

TP183

陈国军, 蔺想红, 王国恩, 王向文. 一种脉冲神经元监督学习的直接计算方法[J]. 电子学报, 2021, 49(2): 331-337.

CHEN Guo-jun, LIN Xiang-hong, WANG Guo-en, WANG Xiang-wen. A Direct Computation Method of Supervised Learning for Spiking Neurons[J]. Acta Electronica Sinica, 2021, 49(2): 331-337.

参考文献

[1] DENG L,WU Y,HU X,et al.Rethinking the performance comparison between SNNS and ANNS neural networks[J].Neural Networks,2020,121:294-307.
[2] BOHTE S M.The evidence for neural information processing with precise spike-times:a survey[J].Natural Computing,2004,3(2):195-206.
[3] BUTTSD A,WENG C,JIN J,et al.Temporal precision in the neural code and the timescales of natural vision[J].Nature,2007,449(7158):92-95.
[4] GERSTNER W,KISTLER W M,et al.Spiking Neuron Models:Single Neurons,Populations,Plasticity[M].Cambridge University Press,2002.
[5] GUTIG R.Spiking neurons can discover predictive features by aggregate-label learning[J].Science,2016,351(6277):4113-4113.
[6] MAASS W.Networks of spiking neurons:The third generation of neural network models[J].Neural Networks,1997,10(9):1659-1671.
[7] 蔺想红,王向文.脉冲神经网络原理及应用[M].北京:科学出版社,2018. LIN Xiang-hong,WANG Xiang-wen.Theory and Application of Spiking Neural Networks[M],Beijing:The Science Publishing Company,2018.(in Chinese)
[8] TAVANAEI A,BELATRECHE A,LI Y,et al.A review of learning in biologically plausible spiking neural networks[J].Neural Networks,2020,122:253-272.
[9] WANG X,LIN X,DANG X.Supervised learning in spiking neural networks:A review of algorithms and evaluations[J].Neural Networks,2020,125:258-280.
[10] KASABOV N,DOBORJEH M G,DOBORJEH Z G.Mapping,learning,visualization,classification,and understanding of fMRI data in the NeuCube evolving spatiotemporal data machine of spiking neural networks[J].IEEE Transactions on Neural Networks and Learning Systems,2017,28(4):887-899.
[11] MORRO A,et al.A stochastic spiking neural network for virtual screening[J].IEEE Transactions on Neural Networks and Learning Systems,2018,29(4):1371-1375.
[12] 蔺想红,王向文,张宁,等.脉冲神经网络的监督学习算法研究综述[J].电子学报,2015,43(3):577-586. LIN Xiang-hong,WANG Xiang-wen,ZHANG Ning,et al.Supervised learning algorithms for spiking neural networks:A review[J].Acta Electronica Sinica,2015,43(3):577-586.(in Chinese)
[13] BOHTE S M,KOK J N,POUTRÉ H L.Error-backpropagation in temporally encoded networks of spiking neurons[J].Neurocomputing,2002,48(1-4):17-37.
[14] GHOSH D S,ADELI H.A new supervised learning algorithm for multiple spiking neural networks with application in epilepsy and seizure detection[J].Neural Networks,2009,22(10):1419-1431.
[15] XU Y,ZENG X,HAN L,et al.A supervised multi-spike learning algorithm based on gradient descent for spiking neural networks[J].Neural Networks,2013,43:99-113.
[16] HEBB D O.The Organization of Behavior A Neuropsychological Theory[M].John Wiley,Chapman & Hall,2013.
[17] CAPORALE N,DAN Y.Spike timing-dependent plasticity:A Hebbian learning rule[J].Annual Review of Neuroscience,2008,31(1):25-46.
[18] WADE J J,MCDAID L J,SANTOS J A,et al.SWAT:A spiking neural network training algorithm for classification problems[J].IEEE Transactions on Neural Networks,2010,21(11):1817-1830.
[19] PONULAK F,KASIŃSKI A.Supervised learning in spiking neural networks with ReSuMe:Sequence learning,classification,and spike shifting[J].Neural Computation,2010,22(22):467-510.
[20] TAHERKHANI A,BELATRECHE A,et al.DL-ReSuMe:A delay learning-based remote supervised method for spiking neurons[J].IEEE Transactions on Neural Networks and Learning Systems,2015,26(12):3137-3149.
[21] LIN X,CHEN G,WANG X,et al.An improved supervised learning algorithm using triplet-based spike-timing-dependent plasticity[A].Proceedings of the 12th International Conference on Intelligent Computing[C].Cham,Switzerland:Springer International Publishing,2016.44-53.
[22] CARNELLA,RICHARDSON D.Linear algebra for times series of spikes[A].Proceedings of the 13th European Symposium on Artificial Neural Networks[C].Evere,Belgium:Dside,2005.363-368.
[23] MOHEMMED A,SCHLIEBS S,et al.Span:Spike pattern association neuron for learning spatio-temporal spike patterns[J].International Journal of Neural Systems,2012,22(4):786-803.
[24] YU Q,TANG H,et al.Precise-spike-driven synaptic plasticity:Learning hetero-association of spatiotemporal spike patterns[J].Plos One,2013,8(11):e78318.
[25] 蔺想红,王向文,党小超.基于脉冲序列核的脉冲神经元监督学习算法[J].电子学报,2016,44(12):2877-2886. LIN Xiang-hong,WANG Xiang-wen,DANG Xiao-chao.A new supervised learning algorithm for spiking neurons based on spike train kernels[J].Acta Electronica Sinica, 2016,44(12):2877-2886.(in Chinese)
[26] WANG X,LIN X,et al.Supervised learning algorithm using for spiking neurons based on nonlinear inner products of spike trains[A].Proceedings of the 12th International Conference on Intelligent Computing[C].Cham,Switzerland:Springer International Publishing,2016.95-104.
[27] LIN X,WANG X,HAO Z.Supervised learning in multilayer spiking neural networks with inner products of spike trains[J].Neurocomputing,2017,237:59-70.
[28] WANG X,LIN X,et al.A delay learning algorithm based on spike train kernels for spiking neurons[J].Frontiers in Neuroence,2010,252(13):1-16.
[29] XU Y,ZENG X,ZHONG S.A new supervised learning algorithm for spiking neurons[J].Neural Computation,2013,25(6):1472-1511.
[30] XIE X,QU H,YI Z,et al.Efficient training of supervised spiking neural network via accurate synaptic-efficiency adjustment method[J].IEEE Transactions on Neural Networks and Learning Systems,2017,28(6):1411-1424.
[31] ZHANG M,QU H,BELATRECHE A,et al.A highly effective and robust membrane potential-driven supervised learning method for spiking neurons[J].IEEE Transactions on Neural Networks and Learning Systems,2019,30(1):123-137.
[32] SCHREIBER S,FELLOUS J M,WHITMER D,et al.A new correlation-based measure of spike timing reliability[J].Neurocomputing,2003,52-54:925-931.

一种脉冲神经元监督学习的直接计算方法

A Direct Computation Method of Supervised Learning for Spiking Neurons

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	江兵兵, 何文达, 吴兴宇, 项俊浩, 洪立斌, 盛伟国. 基于自适应图学习的半监督特征选择[J]. 电子学报, 2022, 50(7): 1643-1652.
[2]	潘敏婷, 王韫博, 朱祥明, 高思宇, 龙明盛, 杨小康. 基于无标签视频数据的深度预测学习方法综述[J]. 电子学报, 2022, 50(4): 869-886.
[3]	张帅, 高旻, 文俊浩, 熊庆宇, 唐旭. 基于自监督学习的去流行度偏差推荐方法[J]. 电子学报, 2022, 50(10): 2361-2371.
[4]	罗会兰, 袁璞, 童康. 基于深度学习的显著性目标检测方法综述[J]. 电子学报, 2021, 49(7): 1417-1427.
[5]	王格格, 郭涛, 余游, 苏菡. 基于生成对抗网络的无监督域适应分类模型[J]. 电子学报, 2020, 48(6): 1190-1197.
[6]	马彬, 王宏明, 谢显中. 基于监督学习的协作宽带压缩频谱检测方案[J]. 电子学报, 2020, 48(12): 2338-2344.
[7]	沈卉卉, 刘国武, 付丽华, 刘智慧, 李宏伟. 一种基于修正动量的RBM算法[J]. 电子学报, 2019, 47(9): 1957-1964.
[8]	余游, 冯林, 王格格, 徐其凤. 一种基于伪标签的半监督少样本学习模型[J]. 电子学报, 2019, 47(11): 2284-2291.
[9]	罗会兰, 张云. 基于深度网络的图像语义分割综述[J]. 电子学报, 2019, 47(10): 2211-2220.
[10]	黄健航, 雷迎科. 基于半监督矩形网络的通信电台个体识别[J]. 电子学报, 2019, 47(1): 1-8.
[11]	刘俊秀, 黄星月, 罗玉玲, 曹弋. 脉冲神经网络硬件互连系统的动态优先级仲裁策略[J]. 电子学报, 2018, 46(8): 1898-1905.
[12]	盛凯, 刘忠, 周德超, 魏启航, 冯成旭. 一种基于证据理论的多类半监督分类算法[J]. 电子学报, 2018, 46(11): 2642-2649.
[13]	周孟, 朱福喜. 基于情感标签的极性分类[J]. 电子学报, 2017, 45(4): 1018-1024.
[14]	罗畅, 王洁, 王鹏飞, 肖军, 肖红. 卷积自编码器中粗粒度池化特征提取研究[J]. 电子学报, 2017, 45(10): 2390-2401.
[15]	刘金平, 陈青, 张进, 唐朝晖. 基于集成学习的交互式图像分割[J]. 电子学报, 2016, 44(7): 1649-1655.