The Kalai-Smorodinski solution for many-objective Bayesian optimization

Abstract : An ongoing scope of research in multi-objective Bayesian optimization is to extend its applicability to a large number of objectives. Recovering the set of optimal compromise solution generally requires lots of observations while being less inter-pretable, since this set tends to grow larger with the number of objectives. We thus propose to focus on a choice of a specific solution originating from game theory, the Kalai-Smorodinski solution, that possesses attractive properties. We further make it insensitive to a monotone transformation of the objectives by considering the objectives in the copula space. A tailored algorithm is proposed to search for the solution, which is tested on a synthetic problem.
Type de document :
Communication dans un congrès
NIPS 2017 - 31st Conference on Neural Information Processing Systems , Dec 2017, Long Beach, United States. pp.1-6
Liste complète des métadonnées

Littérature citée [24 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/hal-01656393
Contributeur : Abderrahmane Habbal <>
Soumis le : mardi 5 décembre 2017 - 15:55:35
Dernière modification le : vendredi 12 janvier 2018 - 01:49:56

Fichier

ks_nips_camera.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-01656393, version 1

Citation

Mickaël Binois, Victor Picheny, Abderrahmane Habbal. The Kalai-Smorodinski solution for many-objective Bayesian optimization. NIPS 2017 - 31st Conference on Neural Information Processing Systems , Dec 2017, Long Beach, United States. pp.1-6. 〈hal-01656393〉

Partager

Métriques

Consultations de la notice

140

Téléchargements de fichiers

72