Non-linear feature extraction by the coordination of mixture models

Abstract : We present a method for non-linear data projection that offers non-linear versions of Principal Component Analysis and Canonical Correlation Analysis. The data is accessed through a probabilistic mixture model only, therefore any mixture model for any type of data can be plugged in. Gaussian mixtures are one example, but mixtures of Bernoulli's to model discrete data might be used as well. The algorithm minimizes an objective function that exhibits one global optimum that can be found by finding the eigenvectors of some matrix. Experimental results on toy data and real data are provided.
Type de document :
Communication dans un congrès
9th Annual Conference of the Advanced School for Computing and Imaging (ASCI '03), Jun 2003, Heijen, Netherlands. pp.287--293, 2003
Liste complète des métadonnées

Littérature citée [15 références]  Voir  Masquer  Télécharger


https://hal.inria.fr/inria-00321490
Contributeur : Jakob Verbeek <>
Soumis le : mardi 8 mars 2011 - 15:04:10
Dernière modification le : lundi 25 septembre 2017 - 10:08:04
Document(s) archivé(s) le : jeudi 9 juin 2011 - 02:45:01

Fichiers

verbeek03asci.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : inria-00321490, version 2

Citation

Jakob Verbeek, Nikos Vlassis, Ben Krose. Non-linear feature extraction by the coordination of mixture models. 9th Annual Conference of the Advanced School for Computing and Imaging (ASCI '03), Jun 2003, Heijen, Netherlands. pp.287--293, 2003. 〈inria-00321490v2〉

Partager

Métriques

Consultations de la notice

223

Téléchargements de fichiers

521