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

Umut Simsekli, Alain Durmus, Roland Badeau, Gaël Richard, Eric Moulines, A. Taylan Cemgil

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.
BibTeX
@inproceedings{simsekli2017parallelized,
  author = {Umut Simsekli and Alain Durmus and Roland Badeau and Gaël Richard and Eric Moulines and A. Taylan Cemgil},
  title = {{Parallelized Stochastic Gradient Markov Chain Monte Carlo algorithms for non-negative matrix factorization}},
  booktitle = {IEEE International Conference on Acoustics, Speech and Signal Processing},
  year = {2017},
  doi = {10.1109/icassp.2017.7952555},
}

Loading related content...

Figure