Local-Meta-Model CMA-ES for Partially Separable Functions

Zyed Bouzarkouna 1, 2 Anne Auger 2 Didier Yu Ding 1
2 TAO - Machine Learning and Optimisation
CNRS - Centre National de la Recherche Scientifique : UMR8623, Inria Saclay - Ile de France, UP11 - Université Paris-Sud - Paris 11, LRI - Laboratoire de Recherche en Informatique
Abstract : In this paper, we propose a new variant of the covariance matrix adaptation evolution strategy with local meta-models (lmm-CMA) for optimizing partially separable functions. We propose to exploit partial separability by building at each iteration a meta-model for each element function (or sub-function) using a full quadratic local model. After introducing the approach we present some first experiments using element functions with dimensions 2 and 4. Our results demonstrate that, as expected, exploiting partial separability leads to an important speedup compared to the standard CMA-ES. We show on the tested functions that the speedup increases with increasing dimensions for a fixed dimension of the element function. On the standard Rosenbrock function the maximum speedup of lambda is reached in dimension 40 using element functions of dimension 2. We show also that higher speedups can be achieved by increasing the population size. The choice of the number of points used to build the meta-model is also described and the computational cost is discussed.
Type de document :
Communication dans un congrès
Genetic and Evolutionary Computation Conference (GECCO 2011), Jul 2011, Dublin, Ireland. pp.869--876, 2011
Liste complète des métadonnées

Littérature citée [14 références]  Voir  Masquer  Télécharger

https://hal.archives-ouvertes.fr/hal-00588977
Contributeur : Anne Auger <>
Soumis le : lundi 2 mai 2011 - 12:41:32
Dernière modification le : jeudi 7 février 2019 - 16:49:46
Document(s) archivé(s) le : jeudi 30 mars 2017 - 09:58:48

Fichier

fp650-bouzarkouna.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-00588977, version 2

Collections

Citation

Zyed Bouzarkouna, Anne Auger, Didier Yu Ding. Local-Meta-Model CMA-ES for Partially Separable Functions. Genetic and Evolutionary Computation Conference (GECCO 2011), Jul 2011, Dublin, Ireland. pp.869--876, 2011. 〈hal-00588977v2〉

Partager

Métriques

Consultations de la notice

692

Téléchargements de fichiers

301