A Component-Wise EM Algorithm for Mixtures

Gilles Celeux 1 Stéphane Chrétien 1, 2 Florence Forbes 1 Abdallah Mkhadri 1
1 IS2 - Statistical Inference for Industry and Health
Inria Grenoble - Rhône-Alpes, LBBE - Laboratoire de Biométrie et Biologie Evolutive
2 NUMOPT - Numerical Optimization
Inria Grenoble - Rhône-Alpes
Abstract : In some situations, EM algorithm shows slow convergence problems. One possible reason is that standard procedures update the parameters simultaneous- ly. In this paper we focus on finite mixture estimation. In this framework, we propose a component-wise EM, which updates the parameters sequentially. We give an interpretation of this procedure as a proximal point algorithm and use it to prove the convergence. Illustrative numerical experiments show how our algorithm compares to EM and a version of the SAGE algorithm.
Type de document :
Rapport
[Research Report] RR-3746, INRIA. 1999
Liste complète des métadonnées

https://hal.inria.fr/inria-00072916
Contributeur : Rapport de Recherche Inria <>
Soumis le : mercredi 24 mai 2006 - 11:16:29
Dernière modification le : jeudi 28 juin 2018 - 14:36:20
Document(s) archivé(s) le : dimanche 4 avril 2010 - 23:27:46

Fichiers

Identifiants

  • HAL Id : inria-00072916, version 1

Collections

Citation

Gilles Celeux, Stéphane Chrétien, Florence Forbes, Abdallah Mkhadri. A Component-Wise EM Algorithm for Mixtures. [Research Report] RR-3746, INRIA. 1999. 〈inria-00072916〉

Partager

Métriques

Consultations de la notice

196

Téléchargements de fichiers

344