On the convergence rates of genetic algorithms

被引：66

作者：

He, J ^{[1
]}

Kang, LS

机构：

[1] No Jiaotong Univ, Dept Comp Sci, Beijing 100044, Peoples R China

[2] Wuhan Univ, State Key Lab Software Engn, Wuhan 430072, Peoples R China

来源：

THEORETICAL COMPUTER SCIENCE | 1999年 / 229卷 / 1-2期

基金：

中国国家自然科学基金; 中国博士后科学基金;

关键词：

genetic algorithms; convergence rate; Markov chain;

D O I：

10.1016/S0304-3975(99)00091-2

中图分类号：

TP301 [理论、方法];

学科分类号：

081202 ;

摘要：

This paper discusses the convergence rates of genetic algorithms by using the minorization condition in the Markov chain theory. We classify genetic algorithms into two kinds: one with time-invariant genetic operators, another with time-variant genetic operators. For the former case, we have obtained the bound on its convergence rate on the general state space; for the later case, we have bounded its convergence rate on the finite state space. (C) 1999 Published by Elsevier Science B.V. All rights reserved.

引用

页码：23 / 39

页数：17