电子学报 ›› 2015, Vol. 43 ›› Issue (3): 577-586.DOI: 10.3969/j.issn.0372-2112.2015.03.024

• 综述评论 • 上一篇    下一篇

脉冲神经网络的监督学习算法研究综述

蔺想红, 王向文, 张宁, 马慧芳   

  1. 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:2014-06-19 修回日期:2014-09-10 出版日期:2015-03-25 发布日期:2015-03-25
  • 作者简介:蔺想红 男,1976年1月生于甘肃天水.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副教授,硕士生导师.研究方向为神经网络、进化计算、人工生命、图像处理. E-mail:linxh@nwnu.edu.cn;王向文 男,1991年3月生于甘肃天水.西北师范大学计算机科学与工程学院硕士研究生.研究方向为神经网络、机器学习. E-mail:wangxiangwen2@163.com
  • 基金资助:

    国家自然科学基金(No.61165002,No.61363058);甘肃省自然科学基金(No.1010RJZA019);甘肃省青年科技基金(No.145RJYA259)

Supervised Learning Algorithms for Spiking Neural Networks:A Review

LIN Xiang-hong, WANG Xiang-wen, ZHANG Ning, MA Hui-fang   

  1. School of Computer Science and Engineering, Northwest Normal University, Lanzhou, Gansu 730070, China
  • Received:2014-06-19 Revised:2014-09-10 Online:2015-03-25 Published:2015-03-25

摘要:

脉冲神经网络是进行复杂时空信息处理的有效工具,但由于其内在的不连续和非线性机制,构建高效的脉冲神经网络监督学习算法非常困难,同时也是该研究领域的重要问题.本文介绍了脉冲神经网络监督学习算法的基本框架,以及性能评价原则,包括脉冲序列学习能力、离线与在线处理性能、学习规则的局部特性和对神经网络结构的适用性.此外,对脉冲神经网络监督学习算法的梯度下降学习规则、突触可塑性学习规则和脉冲序列卷积学习规则进行了详细的讨论,通过对比分析指出现有算法存在的优缺点,并展望了该领域未来的研究方向.

关键词: 脉冲神经网络, 监督学习, 反向传播, 突触可塑性, 卷积

Abstract:

Spiking neural networks are shown to be suitable tools for the processing of spatio-temporal information.However, due to their intricately discontinuous and implicit nonlinear mechanisms, the formulation of efficient supervised learning algorithms for spiking neural networks is difficult, which is an important problem in the research area.In this paper, we introduce the general framework of supervised learning algorithms for spiking neural networks, and analyze their performance evaluations including spike trains learning ability, offline and online processing ability, the locality of learning mechanism and the applicability to network structure.Furthermore, we survey the advance of the research on supervised learning algorithms, which can be divided into three categories according to their differences:gradient descent rule, synaptic plasticity rule, and spike trains convolution rule.Finally, we discuss the advantages and disadvantages of these algorithms, and prospect the problems in current research and some future research directions in this area.

Key words: spiking neural network, supervised learning, backpropagation, synaptic plasticity, convolution

中图分类号: