Contrôle d'un algorithme génétique
Michèle Sébag, Marc Schoenauer · HAL (Le Centre pour la Communication Scientifique Directe) · 1996
This paper briefly recalls the prin1ciple of genetic algorithms (GA) and discusses their efficiency. The Radcliffe' s principles, linking the representation of the problem and the evolution operators, are presented. Last, we address the control of an AG by an induction based approach. Experimental results obtained on typical GA problems and a combinatoric optimization problem, are discussed.