- Recherche et sélection de publications
|
Parallelized Stochastic Gradient Markov Chain Monte Carlo Algorithms for Non-Negative Matrix Factorization
- Umut Simsekli #1, Alain Durmus #1, Roland Badeau #1, Gaël Richard #1, Eric Moulines #2
-
#1 |
Laboratoire traitement et communication de l'information (LTCI)
- Télécm ParisTech
- Institut Mines-Télécom
- Université Paris-Saclay
|
#2 |
Centre de Mathématiques Appliquées - Ecole Polytechnique (CMAP)
- Polytechnique - X
- CNRS : UMR7641
|
- References
- 42nd International Conference on Acoustics, Speech and Signal Processing (ICASSP), New Orleans, USA, IEEE, March 2017,
- Abstract
Stochastic Gradient Markov Chain Monte Carlo (SG-MCMC) methods have become popular in modern data analysis problems due to their computational efficiency. Even though they have proved useful for many statistical models, the application of SG-MCMC to non- negative matrix factorization (NMF) models has not yet been extensively explored. In this study, we develop two parallel SG-MCMC algorithms for a broad range of NMF models. We exploit the conditional independence structure of the NMF models and utilize a stratified sub-sampling approach for enabling parallelization. We illustrate the proposed algorithms on an image restoration task and report encouraging results.
- Keywords
- Stochastic Gradient MCMC, Non-Negative Matrix Factorization, Tweedie Distribution, Beta Divergence, Richardson-Romberg Extrapolation
- Category
- Paper in proceedings
- Research Area(s)
- Engineering Sciences/Signal and Image processing
- Identifier(s)
-
HAL ref. hal-01416357
Bibliographic key US:ICASSP-17
- File(s)
-
- Export
-
- Last update
- on march 20, 2017 by Roland Badeau
|