首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于遗传算法的人工神经网络学习算法
引用本文:李建珍.基于遗传算法的人工神经网络学习算法[J].西北师范大学学报,2002,38(2):33-37.
作者姓名:李建珍
作者单位:西北师范大学教育技术与传播学院 甘肃兰州
摘    要:为了克服和改进BP算法的不足,提出了一种基于遗传算法的神经网络学习算法,仿真结果表明,该算法具有无比的优越性,可避免BP算法易于陷入局部极小值,训练速度慢、误差函数必须可导、受网络结构的限制等缺陷。

关 键 词:权值  阈值  训练速度  遗传算法  人工神经网络学习算法  误差函数  网络结构  BP算法
文章编号:1001-988X(2002)02-0033-05
修稿时间:2001年9月21日

A learning algorithm of artificial neural network based on genetic algorithm
LI Jian-zhen.A learning algorithm of artificial neural network based on genetic algorithm[J].Journal of Northwest Normal University Natural Science (Bimonthly),2002,38(2):33-37.
Authors:LI Jian-zhen
Abstract:In order to get over the insufficiency of BP algorithm, a new learning algorithm of artificial neural network based on genetic algorithm is given. The results of emulation show: using genetic algorithm optimize weight and threshold of artificial neural network, not only can be putted into practice, but has very appealing advantages. The algorithm can get over the insufficiency of BP algorithm, such as: liable to get into local minimum, slow speed in training, derivative error function required, limited to network architecture.
Keywords:artificial neural network  genetic algorithm  weight  threshold  emulation
本文献已被 CNKI 维普 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号