DocumentCode
2850592
Title
Effects of String Length and Mutation Rate on Success Probability of Genetic Algorithm
Author
Zhang, Yu-an ; Sakamoto, Makoto ; Furutani, Hiroshi
Author_Institution
Univ. of Miyazaki, Miyazaki, Japan
Volume
4
fYear
2009
fDate
14-16 Aug. 2009
Firstpage
211
Lastpage
216
Abstract
Mutation plays an important role in the computing of genetic algorithm. In this study, the relationship between success probability and mutation rate is analyzed by using multiplicative landscape. The success probability is defined that there is at least one optimal solution in the stationary state. The effect of mutation was illustrated by studying the distribution of the first order schemata with Markov theory and diffusion equations. This result provides a useful direction on parameter selection.
Keywords
Markov processes; genetic algorithms; Markov theory; diffusion equation; first order schemata; genetic algorithm; multiplicative landscape; mutation rate; parameter selection; string length; success probability; Adaptive control; Cities and towns; Couplings; Frequency; Genetic algorithms; Genetic mutations; Optimal control; Optimization methods; Partial differential equations; Stationary state; Genetic Algorithm; Markov chain; mutation;
fLanguage
English
Publisher
ieee
Conference_Titel
Natural Computation, 2009. ICNC '09. Fifth International Conference on
Conference_Location
Tianjin
Print_ISBN
978-0-7695-3736-8
Type
conf
DOI
10.1109/ICNC.2009.440
Filename
5365360
Link To Document