Preprint -learning unions of orthonormal bases with thresholded singular value decomposition

Sylvain Lesage 1 Rémi Gribonval 1 Frédéric Bimbot 1 Laurent Benaroya 1
1 METISS - Speech and sound data modeling and processing
IRISA - Institut de Recherche en Informatique et Systèmes Aléatoires, Inria Rennes – Bretagne Atlantique
Abstract : We propose a new method to learn overcomplete dictionaries for sparse coding. The method is designed to learn dictionaries structured as unions of orthonormal bases. The interest of such a structure is manifold. Indeed, it seems that many signals or images can be modeled as the super-imposition of several layers with sparse decompositions in as many bases. Moreover, in such dictionaries, the efficient Block Coordinate Relaxation (BCR) algorithm can be used to compute sparse decompositions. We show that it is possible to design an iterative learning algorithm that produces a dictionary with the required structure. Each step is based on the coefficients estimation, using a variant of BCR, followed by the update of one chosen basis, using Singular Value Decomposition. We assess experimentally how well the learning algorithm recovers dictionaries that may or may not have the required structure, and to what extent the noise level is a disturbing factor.
Type de document :
Pré-publication, Document de travail
This is the extended version of the article with the same title published at ICASSP 2005. 2004
Liste complète des métadonnées

Littérature citée [11 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/hal-01637825
Contributeur : Rémi Gribonval <>
Soumis le : samedi 18 novembre 2017 - 08:54:38
Dernière modification le : mercredi 16 mai 2018 - 11:23:03
Document(s) archivé(s) le : lundi 19 février 2018 - 12:31:10

Fichier

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

Identifiants

  • HAL Id : hal-01637825, version 1

Collections

Citation

Sylvain Lesage, Rémi Gribonval, Frédéric Bimbot, Laurent Benaroya. Preprint -learning unions of orthonormal bases with thresholded singular value decomposition. This is the extended version of the article with the same title published at ICASSP 2005. 2004. 〈hal-01637825〉

Partager

Métriques

Consultations de la notice

653

Téléchargements de fichiers

30