Noiseless Independent Factor Analysis with mixing constraints in a semi-supervised framework. Application to railway device fault diagnosis.

Etienne Côme 1 Latifa Oukhellou 2 Patrice Aknin 2 Thierry Denoeux 3
1 SAMOS - Statistique Appliquée et MOdélisation Stochastique
2 INRETS/LTN - Laboratoire des Technologies Nouvelles
3 Heudiasyc - Heuristique et Diagnostic des Systèmes Complexes [Compiègne]
Abstract : In Independent Factor Analysis (IFA), latent components (or sources) are recovered from only their linear observed mixtures. Both the mixing process and the source densities (that are assumed to be gener- ated according to mixtures of Gaussians) are learned from observed data. This paper investigates the possibility of estimating the IFA model in its noiseless setting when two kinds of prior information are incorporated: constraints on the mixing process and partial knowledge on the cluster membership of some examples. Semi-supervised or partially supervised learning frameworks can thus be handled. These two proposals have been initially motivated by a real-world application that concerns fault diag- nosis of a railway device. Results from this application are provided to demonstrate the ability of our approach to enhance estimation accuracy and remove indeterminacy commonly encountered in unsupervised IFA such as source permutations.
Type de document :
Communication dans un congrès
International Conference on Artificial Neural Networks (ICANN),, Sep 2009, Limassol, Cyprus. 5769, pp.416-425, 2007, Lecture Notes in Computer Science. 〈10.1007/978-3-642-04277-5_42〉
Liste complète des métadonnées

Littérature citée [15 références]  Voir  Masquer  Télécharger
https://hal-paris1.archives-ouvertes.fr/hal-00446628
Contributeur : Etienne Côme <>
Soumis le : mercredi 13 janvier 2010 - 12:25:41
Dernière modification le : mardi 10 avril 2018 - 11:02:02
Document(s) archivé(s) le : jeudi 17 juin 2010 - 22:43:06

Fichier

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

Identifiants

Collections

UNIV-PARIS1 | SAMOS | IFSTTAR | HEUDIASYC

Citation

Etienne Côme, Latifa Oukhellou, Patrice Aknin, Thierry Denoeux. Noiseless Independent Factor Analysis with mixing constraints in a semi-supervised framework. Application to railway device fault diagnosis.. International Conference on Artificial Neural Networks (ICANN),, Sep 2009, Limassol, Cyprus. 5769, pp.416-425, 2007, Lecture Notes in Computer Science. 〈10.1007/978-3-642-04277-5_42〉. 〈hal-00446628〉

Exporter

BibTeX TEI DC DCterms EndNote

Partager

Métriques

Consultations de la notice

259

Téléchargements de fichiers

113

Contact

support.ccsd.cnrs.fr
hal.support@ccsd.cnrs.fr
Logo CCSD