October, Tuesday 24th

14:30 (Shannon amphitheatre, building 660) (see location):

Benjamin Guedj

(MODAL project-team of Inria Lille [MOdels for Data Analysis and Learning] / Laboratoire Paul Painlevé, University of Lille)

Title: A quasi-Bayesian perspective to NMF: theory and applications

Abstract:

Quasi-Bayesian estimators are increasingly popular in statistics and machine learning, due to their generalization properties and flexibility. In a recent work (Alquier & Guedj 2017, Mathematical Methods of Statistics), we have proposed a quasi-Bayesian estimator for non-negative matrix factorization. I will present a quick overview of quasi- and PAC-Bayesian frameworks and discuss our theoretical and algorithmic contributions. A short demo of our method for digits recognition will conclude the talk.

Reference:

http://dx.doi.org/10.3103/S1066530717010045

Website:

https://bguedj.github.io




Contact: guillaume.charpiat at inria.fr