Title :
Adapting mutations in genetic algorithms using gene flow principles
Author :
Greenwood, Garrison W.
Author_Institution :
Dept. of Electr. & Comput. Eng., Portland State Univ., USA
Abstract :
Bit mutation in genetic algorithms is usually done with a single fixed probability. Methods to adapt this probability have been suggested, but they operate at the genome level. This paper describes a gene level adaption scheme, based on allele frequencies, which gives a better escape from local optima.
Keywords :
genetic algorithms; probability; allele frequencies; bit mutation; gene flow principles; gene level adaption scheme; genetic algorithms; Bioinformatics; Biological cells; Conducting materials; Frequency; Genetic algorithms; Genetic engineering; Genetic mutations; Genomics; Sampling methods;
Conference_Titel :
Evolutionary Computation, 2003. CEC '03. The 2003 Congress on
Print_ISBN :
0-7803-7804-0
DOI :
10.1109/CEC.2003.1299833
