Share Email Print
cover

Proceedings Paper

Self-adaptive parameters in genetic algorithms
Author(s): Eric Pellerin; Luc Pigeon; Sylvain Delisle
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

Genetic algorithms are powerful search algorithms that can be applied to a wide range of problems. Generally, parameter setting is accomplished prior to running a Genetic Algorithm (GA) and this setting remains unchanged during execution. The problem of interest to us here is the self-adaptive parameters adjustment of a GA. In this research, we propose an approach in which the control of a genetic algorithm’s parameters can be encoded within the chromosome of each individual. The parameters’ values are entirely dependent on the evolution mechanism and on the problem context. Our preliminary results show that a GA is able to learn and evaluate the quality of self-set parameters according to their degree of contribution to the resolution of the problem. These results are indicative of a promising approach to the development of GAs with self-adaptive parameter settings that do not require the user to pre-adjust parameters at the outset.

Paper Details

Date Published: 12 April 2004
PDF: 12 pages
Proc. SPIE 5433, Data Mining and Knowledge Discovery: Theory, Tools, and Technology VI, (12 April 2004); doi: 10.1117/12.542156
Show Author Affiliations
Eric Pellerin, Defence R&D Canada Valcartier (Canada)
Université du Québec à Trois-Rivières (Canada)
Luc Pigeon, Defence R&D Canada Valcartier (Canada)
Sylvain Delisle, Univ. du Quebec a Trois-Rivieres (Canada)


Published in SPIE Proceedings Vol. 5433:
Data Mining and Knowledge Discovery: Theory, Tools, and Technology VI
Belur V. Dasarathy, Editor(s)

© SPIE. Terms of Use
Back to Top