摘要
本文讨论了进化神经网络的编码表示机制,分析了它们的优缺点;提出了遗传算法的一种图文法编码表示机制,给出了相应的算子定义,以及模式、模式长度及其阶的定义;证明了一个基于图文法表示机制的遗传算法模式定理,描述了交叉和突变对模式作用的效果。
This paper discusses the encoding representations of evolving neural networks,analyses the advantages and disadvantages of these methods.It presents an encoding representation of genetic algorithms based on graph grammar,and gives the corresponding definitions of genetic operators,schema,schema order,and length as well.It proves a schema theorem for genetic algorithms in which representation schema is based on graph grammar.The effect of crossover and mutation on schemata is described.
出处
《计算机工程与科学》
CSCD
1998年第4期11-16,共6页
Computer Engineering & Science
基金
邮电部切块经费资助
关键词
遗传算法
图文法
进化神经网络
genetic algorithms,graph grammar,evolving neural network.