A BAYESIAN NON PARAMETRIC APPROACH TO LEARN ...
Document type :
Communication dans un congrès avec actes
Title :
A BAYESIAN NON PARAMETRIC APPROACH TO LEARN DICTIONARIES WITH ADAPTED NUMBERS OF ATOMS
Author(s) :
Dang, Hong-Phuong [Auteur]
Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 [CRIStAL]
Chainais, Pierre [Auteur]
Centrale Lille
Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 [CRIStAL]
Chainais, Pierre [Auteur]

Centrale Lille
Conference title :
IEEE Workshop on Machine Learning for Signal Processing MLSP'2015
City :
Boston
Country :
Etats-Unis d'Amérique
Start date of the conference :
2015-09-17
Book title :
Proceedings of IEEE Workshop on Machine Learning for Signal Processing
Publication date :
2015
English keyword(s) :
Index Terms— sparse representations
dictionary learning
inverse problems
Indian Buffet Process
dictionary learning
inverse problems
Indian Buffet Process
HAL domain(s) :
Sciences de l'ingénieur [physics]/Traitement du signal et de l'image [eess.SP]
Statistiques [stat]/Applications [stat.AP]
Informatique [cs]/Apprentissage [cs.LG]
Informatique [cs]/Traitement du signal et de l'image [eess.SP]
Statistiques [stat]/Applications [stat.AP]
Informatique [cs]/Apprentissage [cs.LG]
Informatique [cs]/Traitement du signal et de l'image [eess.SP]
English abstract : [en]
Learning redundant dictionaries for sparse representation from sets of patches has proven its efficiency in solving inverse problems. In many methods, the size of the dictionary is fixed in advance. Moreover the optimization ...
Show more >Learning redundant dictionaries for sparse representation from sets of patches has proven its efficiency in solving inverse problems. In many methods, the size of the dictionary is fixed in advance. Moreover the optimization process often calls for the prior knowledge of the noise level to tune parameters. We propose a Bayesian non parametric approach which is able to learn a dictionary of adapted size : the adequate number of atoms is inferred thanks to an Indian Buffet Process prior. The noise level is also accurately estimated so that nearly no parameter tuning is needed. Numerical experiments illustrate the relevance of the resulting dictionaries.Show less >
Show more >Learning redundant dictionaries for sparse representation from sets of patches has proven its efficiency in solving inverse problems. In many methods, the size of the dictionary is fixed in advance. Moreover the optimization process often calls for the prior knowledge of the noise level to tune parameters. We propose a Bayesian non parametric approach which is able to learn a dictionary of adapted size : the adequate number of atoms is inferred thanks to an Indian Buffet Process prior. The noise level is also accurately estimated so that nearly no parameter tuning is needed. Numerical experiments illustrate the relevance of the resulting dictionaries.Show less >
Language :
Anglais
Peer reviewed article :
Oui
Audience :
Internationale
Popular science :
Non
Comment :
Best Paper Award
Collections :
Source :
Files
- https://hal.archives-ouvertes.fr/hal-01249704/document
- Open access
- Access the document
- https://hal.archives-ouvertes.fr/hal-01249704/document
- Open access
- Access the document
- https://hal.archives-ouvertes.fr/hal-01249704/document
- Open access
- Access the document
- document
- Open access
- Access the document
- final26.pdf
- Open access
- Access the document