Ensemble Learning for Free with Evolutionary Algorithms ?

Abstract : Evolutionary Learning proceeds by evolving a population of classifiers, from which it generally returns (with some notable exceptions) the single best-of-run classifier as final result. In the meanwhile, Ensemble Learning, one of the most efficient approaches in supervised Machine Learning for the last decade, proceeds by building a population of diverse classifiers. Ensemble Learning with Evolutionary Computation thus receives increasing attention. The Evolutionary Ensemble Lear\-ning (EEL) approach presented in this paper features two contributions. First, a new fitness function, inspired by co-evolution and enforcing the classifier diversity, is presented. Further, a new selection criterion based on the classification margin is proposed. This criterion is used to extract the classifier ensemble from the final population only (Off-line) or incrementally along evolution (On-line). Experiments on a set of benchmark problems show that Off-line outperforms single-hypothesis evolutionary learning and state-of-art Boosting and generates smaller classifier ensembles.
Type de document :
Communication dans un congrès
Dirk Thierens et al. GECCO, Jul 2007, London, United Kingdom. ACM, pp.1782-1789, 2007
Liste complète des métadonnées

Littérature citée [24 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/inria-00144010
Contributeur : Marc Schoenauer <>
Soumis le : lundi 30 avril 2007 - 10:35:17
Dernière modification le : mercredi 28 novembre 2018 - 15:36:02
Document(s) archivé(s) le : mercredi 7 avril 2010 - 01:16:50

Fichiers

eel-gecco.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : inria-00144010, version 1
  • ARXIV : 0704.3905

Collections

Citation

Christian Gagné, Michèle Sebag, Marc Schoenauer, Marco Tomassini. Ensemble Learning for Free with Evolutionary Algorithms ?. Dirk Thierens et al. GECCO, Jul 2007, London, United Kingdom. ACM, pp.1782-1789, 2007. 〈inria-00144010〉

Partager

Métriques

Consultations de la notice

461

Téléchargements de fichiers

867