Machine Learning Seminar

Florent Krzakala, Institut Universitaire de France
Structured Low Rank Matrix Factorization

A large amount of interesting problems in machine learning and statistics can be expressed as a low-rank structured matrix factorization problem, such as sparse PCA, planted clique, sub-matrix localization, clustering of mixtures of Gaussians or community detection in a graph. I will discuss how recent ideas in statistical physics and information theory have led, on the one hand, to new mathematical insights in these problems, leading to a characterization of the optimal possible performances and the mutual information in the Bayes-optimal scenario, and on the other to the development of new powerful algorithms. http://krzakala.org/