A Study of the Effects of Dimensionality on Stochastic Hill Climbers and Estimation of Distribution Algorithms

Abstract : One of the most important features of an optimization method is its response to an increase in the number of variables, n . Random stochastic hill climber (SHC) and univariate marginal distribution algorithms (UMDA) are two fundamentally different stochastic optimizers. SHC proceeds with local perturbations while UMDA infers and uses a global probability density. The response to dimensionality of the two methods is compared both numerically and theoretically on unimodal functions. SHC response is , while UMDA response ranges from to . On two test problems whose sizes go up to 7 200 , SHC is faster than UMDA.
Type de document :
Chapitre d'ouvrage
Liardet, Pierre Editor: Collet, Pierre : Fonlupt, Cyril : Lutton, Evelyne : Schoenauer, Marc. Artificial Evolution, Springer Berlin / Heidelberg, p 27 - 38, 2004, Lecture Notes in Computer Science, 978-3-540-21523-3. 〈10.1007/978-3-540-24621-3_3〉
Liste complète des métadonnées

https://hal-emse.ccsd.cnrs.fr/emse-00686907
Contributeur : Florent Breuil <>
Soumis le : mercredi 11 avril 2012 - 15:53:58
Dernière modification le : mardi 17 octobre 2017 - 12:08:01

Identifiants

Citation

Laurent Grosset, Rodolphe Le Riche, Raphael Haftka. A Study of the Effects of Dimensionality on Stochastic Hill Climbers and Estimation of Distribution Algorithms. Liardet, Pierre Editor: Collet, Pierre : Fonlupt, Cyril : Lutton, Evelyne : Schoenauer, Marc. Artificial Evolution, Springer Berlin / Heidelberg, p 27 - 38, 2004, Lecture Notes in Computer Science, 978-3-540-21523-3. 〈10.1007/978-3-540-24621-3_3〉. 〈emse-00686907〉

Partager

Métriques

Consultations de la notice

100