DocumentCode :
1652052
Title :
Local convergence rate of evolutionary algorithm with combined mutation operator
Author :
Nam Geun Kim ; Won, Jin M. ; Lee, Jin S. ; Kim, Nam Geun
Author_Institution :
Div. of Electr. & Comput. Eng., Pohang Univ. of Sci. & Technol., South Korea
Volume :
1
fYear :
2002
Firstpage :
261
Abstract :
An appropriate mutation operator of the evolutionary algorithm (EA) maintains a balance between exploration and exploitation. This balance is usually satisfied by using the combined mutation operators (CMOs) of the Gaussian and Cauchy random variables. This paper studies the convergence property of the CMO. As a good model of the CMO, it proposes to use the decision factor /spl alpha/, the probability of choosing the Gaussian random variable between the Gaussian and Cauchy random variables for a mutation operator. This paper shows that the optimal convergence rate and the associated optimal mutation step size are monotonically decreasing with respect to /spl alpha/.
Keywords :
convergence; evolutionary computation; probability; Cauchy random variables; Gaussian random variables; combined mutation operator; decision factor; evolutionary algorithm; local convergence rate; optimal convergence rate; optimal mutation step size; probability; Appropriate technology; CMOS technology; Convergence; Evolutionary computation; Genetic mutations; Information geometry; Maintenance engineering; Random variables; Robustness; Semiconductor device modeling;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Evolutionary Computation, 2002. CEC '02. Proceedings of the 2002 Congress on
Conference_Location :
Honolulu, HI, USA
Print_ISBN :
0-7803-7282-4
Type :
conf
DOI :
10.1109/CEC.2002.1006244
Filename :
1006244
Link To Document :
بازگشت