article · 1997

Adaptive Genetic Operators in Elimination Genetic Algorithm

Domagoj Jakobovic

Cited by 0Abstract and details only

Abstract

This paper describes a variant of Genetic Algorithm (GA) with adaptive probabilities of crossover and mutation. The application of GA presented here refers to multimodal function optimization and some deceptive problems. Adaptive probabilities do not, however, restrict the applicability in any way. The adaptive method is designed for steady-state GA with elimination selection. The main idea of the method lies in selective choosing of solutions to be crossed and in varying number of mutations. The adaptive GA is analyzed concerning its capability of locating the global optimum in a multimodal landscape. Several test functions with various degrees of complexity, employed earlier for comparative studies, have been optimized. The performance of standard steady-state GA is compared with that of the adaptive GA. A notable improvement can be perceived in the results: the adaptive GA converges faster and gets stuck at a local optimum fewer times. There is also less preparation needed for the user to get the algorithm to work, as it calculates some of its control parameters itself.

Cite this paper

APA 7

Jakobovic, D. (1997). Adaptive Genetic Operators in Elimination Genetic Algorithm.

MLA 9

Jakobovic, Domagoj. “Adaptive Genetic Operators in Elimination Genetic Algorithm.” 1997.

Chicago

Jakobovic, Domagoj. 1997. “Adaptive Genetic Operators in Elimination Genetic Algorithm.”

Harvard

Jakobovic, D. (1997) ‘Adaptive Genetic Operators in Elimination Genetic Algorithm’.

BibTeX

@article{jakobovic1997adaptive,
  title = {{Adaptive Genetic Operators in Elimination Genetic Algorithm}},
  author = {Jakobovic, Domagoj},
  year = {1997}
}

Made from the index record. Check names, capitalisation and page numbers against the paper. Need another style? AI for Research has over 10,000.