Title :

An Oracle Inequality for Quasi-Bayesian Non-Negative Matrix Factorization

Author(s) :

Alquier, Pierre [Auteur]
Guedj, Benjamin [Auteur]
Laboratoire Paul Painlevé - UMR 8524 [LPP]

Journal title :

Mathematical Methods of Statistics

Abbreviated title :

MMS

Volume number :

26

Pages :

55-67

Publisher :

Allerton Press, Springer (link)

Publication date :

2017

ISSN :

1066-5307

Keyword(s) :

Non-negative matrix factorization
Gibbs sampler
PAC-Bayesian theory
Blockwise coordinate optimization
Oracle inequality

HAL domain(s) :

Statistiques [stat]/Machine Learning [stat.ML]

English abstract : [en]

The aim of this paper is to provide some theoretical understanding of Bayesian non-negative matrix factorization methods. We derive an oracle inequality for a quasi-Bayesian estimator. This result holds for a very general ...
Show more >The aim of this paper is to provide some theoretical understanding of Bayesian non-negative matrix factorization methods. We derive an oracle inequality for a quasi-Bayesian estimator. This result holds for a very general class of prior distributions and shows how the prior affects the rate of convergence. We illustrate our theoretical results with a short numerical study along with a discussion on existing implementations .Show less >

Language :

Anglais

Audience :

Internationale

Popular science :

Non

Administrative institution(s) :

CNRS
Université de Lille

Collections :

• METRICS : Evaluation des technologies de santé et des pratiques médicales - ULR 2694

Submission date :

2020-06-08T14:11:20Z
2020-06-09T09:15:12Z