Parallelized Stochastic Gradient Markov Chain Monte Carlo algorithms for non-negative matrix factorization

Umut Simsekli, Alain Durmus, Roland Badeau, Gael Richard, Eric Moulines, A. Taylan Cemgil
2017 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)  
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 nonnegative 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
more » ... ilize a stratified sub-sampling approach for enabling parallelization. We illustrate the proposed algorithms on an image restoration task and report encouraging results.
doi:10.1109/icassp.2017.7952555 dblp:conf/icassp/SimsekliDBRMC17 fatcat:362nxrdr7fghfl572uuej4pkji