SAGA: Sparse And Geometry-Aware non-negative matrix factorization through non-linear local embedding

Abstract : This paper presents a new non-negative matrix factorization technique which (1) allows the decomposition of the original data on multiple latent factors accounting for the geometrical structure of the manifold embedding the data; (2) provides an optimal representation with a controllable level of sparsity; (3) has an overall linear complexity allowing handling in tractable time large and high dimensional datasets. It operates by coding the data with respect to local neighbors with non-linear weights. This locality is obtained as a consequence of the simultaneous sparsity and convexity constraints. Our method is demonstrated over several experiments, including a feature extraction and classification task, where it achieves better performances than the state-of-the-art factorization methods, with a shorter computational time.
Type de document :
Article dans une revue
Machine Learning, Springer Verlag, 2014, pp.1--23
Liste complète des métadonnées


https://hal.archives-ouvertes.fr/hal-01018683
Contributeur : Nicolas Courty <>
Soumis le : vendredi 4 juillet 2014 - 16:40:02
Dernière modification le : mercredi 2 août 2017 - 10:10:18
Document(s) archivé(s) le : samedi 4 octobre 2014 - 13:00:14

Fichier

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

Identifiants

  • HAL Id : hal-01018683, version 1

Citation

Nicolas Courty, Xing Gong, Jimmy Vandel, Thomas Burger. SAGA: Sparse And Geometry-Aware non-negative matrix factorization through non-linear local embedding. Machine Learning, Springer Verlag, 2014, pp.1--23. <hal-01018683>

Partager

Métriques

Consultations de
la notice

9209

Téléchargements du document

578