基于STDP规则的脉冲神经网络研究

doi:10.19678/j.issn.1000-3428.0055311

摘要/Abstract

摘要： 人类对于生物系统信息的处理主要依赖于构成复杂神经网络的数十亿个神经元，并且信息以脉冲的形式进行传输。利用STDP学习算法构建基于LIF模型的两层脉冲神经网络结构，并对分类层算法进行改进，提出一种投票竞争机制。通过多次训练后对神经元表现类别进行竞争投票，优化同等神经元数量的网络机构在图像分类问题中的性能。在MNIST数据集上进行实验验证，结果表明，该投票竞争机制准确率达到98.1%，与同等网络规模下未采用投票竞争机制的脉冲神经网络相比，准确率平均提高了约6%，而且当神经元数目较少时，在不增加训练时间情况下，可以取得与更加复杂网络结构相同的训练结果。

关键词: STDP规则, 脉冲神经网络, LIF模型, 投票竞争机制, 图像识别

Abstract: The way human beings process information mainly relies on billions of neurons that constitute a complex neural network,and the information is transmitted in the form of pulses.The STDP learning algorithm is used to construct a two-layer Spiking Neural Network(SNN) structure based on the LIF model,and a voting competition mechanism is proposed based on the improved classification layer algorithm.Through the competitive voting of neuron performance categories after multiple times of training,the performance of the network organization with the same number of neurons on the image classification problem is optimized.Results of experimental verification on the MNIST data set show that the accuracy rate of the voting competition mechanism reaches 98.1%,about 6%on average higher than that of the SNN under the same network scale without the voting competition mechanism.When the number of neurons is small,the voting competition mechanism can achieve the same training results as more complex network structures do without increasing the training time.

Key words: STDP rule, Spiking Neural Network(SNN), LIF model, voting competition mechanism, image recognition

中图分类号:

TP18

庄祖江, 房玉, 雷建超, 刘栋博, 王海滨. 基于STDP规则的脉冲神经网络研究[J]. 计算机工程, 2020, 46(9): 83-88,94.

ZHUANG Zujiang, FANG Yu, LEI Jianchao, LIU Dongbo, WANG Haibin. Research on Spiking Neural Network Based on STDP Rule[J]. Computer Engineering, 2020, 46(9): 83-88,94.

https://www.ecice06.com/CN/Y2020/V46/I9/83

图/表 8

20200915134804

20200915134807

20200915134812

20200915134815

20200915134818

20200915134821

20200915134824

20200915134827

参考文献

[1] PAUGAM-MOISY H,BOHTE S.Computing with spiking neuron networks[C]//Proceedings of HNC'12.Berlin,Germany:Springer,2012:335-376.
[2] MAASS W.Networks of spiking neurons:the third generation of neural network models[J].Neural Networks,1997,10(9):1659-1671.
[3] MAAS W.Noisy spiking neurons with temporal coding have more computational power than sigmoidal neurons[J].Advances in Neural Information Processing Systems,1997,9:211-217.
[4] GOODMAN D F,BRETTE R.Brian:a simulator for spiking neural networks in python[J].Frontiers in Neuroinformatics,2008,2:5-10.
[5] GOODMAN D F,BRETTE R.The brian simulator[J].Frontiers in Neuroscience,2009,3:26-31.
[6] MAASS W,NATSCHLÄGER T,MARKRAM H.Real-time computing without stable states:a new framework for neural computation based on perturbations[J].Neural Computation,2002,14(11):2531-2560.
[7] BUONOMANO D V,MAASS W.State-dependent computations:spatiotemporal processing in cortical networks[J].Nature Reviews Neuroscience,2009,10(2):113.
[8] MASUTA H,KUBOTA N.The perception for partner robot using spiking neural network in dynamic environment[C]//Proceedings of SICE'08.Washington D.C.,USA:IEEE Press,2008:2001-2006.
[9] GERSTNER W,KISTLER W M.Spiking neuron models:single neurons,populations,plasticity[M].Cambridge,UK:Cambridge University Press,2002.
[10] LIN Xianghong,ZHANG Ning,CUI Wenbo,et al.Image segmentation method of spiking neural network based on TTFS coding[J].Computer Engineering,2015,41(8):227-232.(in Chinese)蔺想红,张宁,崔文博,等.基于TTFS编码的脉冲神经网络图像分割方法[J].计算机工程,2015,41(8):227-232.
[11] LIN Xianghong,LI Dan,WANG Xiangwen,et al.Online supervised learning algorithm for spiking neuron based on spiking sequence composite kernel[J].Computer Engineering,2017,43(12):197-202.(in Chinese)蔺想红,李丹,王向文,等.基于脉冲序列合成核的脉冲神经元在线监督学习算法[J].计算机工程,2017,43(12):197-202.
[12] WANG Lei,WANG Lianming.An improved self-organizing map spiking neural networks based on STDP rule[J].Journal of Northeast Normal University(Natural Science Edition),2017,49(3):52-56.(in Chinese)王蕾,王连明.一种改进的基于STDP规则的SOM脉冲神经网络[J].东北师大学报(自然科学版),2017,49(3):52-56.
[13] LIU Junxiu,HUANG Xingyue,LUO Yuling,et al.Dynamicpriority arbitration strategy for interconnections of hardware spiking neural networks[J].Acta Electronica Sinica,2018,46(8):1898-1905.(in Chinese)刘俊秀,黄星月,罗玉玲,等.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.
[14] MARKRAM H,LÜBKE J,FROTSCHER M,et al.Regulation of synaptic efficacy by coincidence of postsynaptic APs and EPSPs[J].Science,1997,275(5297):213-215.
[15] KRAUSE T U.Rate coding and temporal coding in a neural network[EB/OL].[2019-05-20].http://www.ini.rub.de/PEOPLE/rolf/.
[16] PONULAK F,KASINSKI A.Supervised learning in spiking neural networks with ReSuMe:sequence learning,classification,and spike shifting[J].Neural Computation,2010,22(2):467-510.
[17] BOHTE S M,KOK J N,LA POUTRÉ J A.SpikeProp:backpropagation for networks of spiking neurons[C]//Proceedings of ESANN'00.Bruges,Belgium:[s.n.],2000:17-37.
[18] KEMPTER R,GERSTNER W,VAN HEMMEN J L.Hebbian learning and spiking neurons[J].Physical Review E,1999,59(4):4498.
[19] CHENG Long,LIU Yang.Spiking neural networks:model,learning algorithms and applications[J].Control and Decision,2018,33(5):923-937.(in Chinese)程龙,刘洋.脉冲神经网络:模型、学习算法与应用[J].控制与决策,2018,33(5):923-937.
[20] MAASS W.Analog computations on networks of spiking neurons[C]//Proceedings of Italian Workshop on Neural Nets.[S.1.]:World Scientific Publishing,1996:99-111.
[21] FENG J,BROWN D.Integrate-and-fire models with nonlinear leakage[J].Bulletin of Mathematical Biology,2000,62(3):467-481.
[22] XU Yan,YANG Jing.Research on affect factors of spiking neuron sequence learning method[J].Computer Engineering,2015,41(11):194-201.(in Chinese)徐彦,杨静.脉冲神经元序列学习方法的影响因素研究[J].计算机工程,2015,41(11):194-201.
[23] GILSON M,MASQUELIER T,HUGUES E.STDP allows fast rate-modulated coding with poisson-like spike trains[J].PLoS Computational Biology,2011,7(10):100-111.
[24] JUG F.On competition and learning in cortical structures[D].Zurich,Switzerland:ETH Zurich,2012.
[25] HUSSAIN S,LIU S C,BASU A.Improved margin multi-class classification using dendritic neurons with morphological learning[C]//Proceedings of 2014 IEEE International Symposium on Circuits and Systems.Washington D.C.,USA:IEEE Press,2014:2640-2643.
[26] O'CONNOR P,NEIL D,LIU S C,et al.Real-time classification and sensor fusion with a spiking deep belief network[J].Frontiers in Neuroscience,2013,7:178-188.
[27] ZHAO B,DING R,CHEN S,et al.Feedforward categorization on AER motion events using cortex-like features in a spiking neural network[J].IEEE Transactions on Neural Networks and Learning Systems,2014,26(9):1963-1978.
[28] DIEHL P U,COOK M.Unsupervised learning of digit recognition using spike-timing-dependent plasticity[J].Frontiers in Computational Neuroscience,2015,9:23-35.

选择文件类型/文献管理软件名称

选择包含的内容