Genetic algorithm involving coevolution mechanism to search for effective genetic information

Author

Handa, Hisashi ; Baba, Noria ; Katai, Osamu ; Sawaragi, Tetsuo ; Horiuchi, Tadashi

Author_Institution

Dept. of Precision Eng., Kyoto Univ., Japan

fYear

1997

fDate

13-16 Apr 1997

Firstpage

709

Lastpage

714

Abstract

A new genetic algorithm which exploits an idea of “coevolution” is proposed. The proposed method consists of two GAs: Host GA and Parasite GA. The Host GA searches for the solutions, and these two GAs are closely related to each other. The Parasite GA plays an important role in searching for useful schemata in the Host GA. Furthermore, two methods of fitness evaluation of Parasite GA are examined: differentiating method and averaging method. The differentiating method will yield the search for schemata that are not yet discovered by the Host GA. The averaging method will yield the search for schemata that have high average of fitness. Various computer simulations confirm the effectiveness of the proposed methods

Keywords

genetic algorithms; search problems; stochastic processes; Host GA; Parasite GA; averaging method; coevolution; coevolution mechanism; computer simulations; differentiating method; effective genetic information searching; fitness evaluation; genetic algorithm; Art; Biological cells; Cities and towns; Computer simulation; Cultural differences; Evolutionary computation; Genetic algorithms; Precision engineering; Search methods; Stochastic processes;

fLanguage

English

Publisher

ieee

Conference_Titel

Evolutionary Computation, 1997., IEEE International Conference on

Conference_Location

Indianapolis, IN

Print_ISBN

0-7803-3949-5

Type

conf

DOI

10.1109/ICEC.1997.592427

Filename

592427