A comparison of an estimation of distribution algorithm and a stochastic hill-climber for composite optimization problems

Abstract : Evolutionary algorithms (EA) have become a standard tool for the optimization of complex composite structures because of their ability to solve combinatorial problems. However, several studies have shown that simpler algorithms, such as stochastic hill climbers (SHC) can be more e cient even on problems designed to demonstrate EAs superiority, such as the Royal Road problem. The present paper compares the performance of a variant of EA, the univariate marginal distribution algorithm (UMDA) with that of an SHC on di erent tness landscapes found in laminate optimization problems and identi es factors that in uence the algorithms' relative performance. In particular, it is found that mUMDA, a hybrid algorithm that combines UMDA's global distribution learning and SHC's local random search, outperforms SHC on large, highly constrained problems and on multimodal problems.
Type de document :
Communication dans un congrès
18th Annual Technical Conference American Society for Composites, Oct 2003, Gainesville, United States. paper 168, 2003
Liste complète des métadonnées

https://hal-emse.ccsd.cnrs.fr/emse-00687059
Contributeur : Florent Breuil <>
Soumis le : jeudi 12 avril 2012 - 10:13:16
Dernière modification le : mardi 23 octobre 2018 - 14:36:09

Identifiants

  • HAL Id : emse-00687059, version 1

Citation

Laurent Grosset, Rodolphe Le Riche, Raphael Haftka. A comparison of an estimation of distribution algorithm and a stochastic hill-climber for composite optimization problems. 18th Annual Technical Conference American Society for Composites, Oct 2003, Gainesville, United States. paper 168, 2003. 〈emse-00687059〉

Partager

Métriques

Consultations de la notice

88