Mixed Membership Matrix Factorization

11 years 23 days ago
Mixed Membership Matrix Factorization
Discrete mixed membership modeling and continuous latent factor modeling (also known as matrix factorization) are two popular, complementary approaches to dyadic data analysis. In this work, we develop a fully Bayesian framework for integrating the two approaches into unified Mixed Membership Matrix Factorization (M3 F) models. We introduce two M3 F models, derive Gibbs sampling inference procedures, and validate our methods on the EachMovie, MovieLens, and Netflix Prize collaborative filtering datasets. We find that, even when fitting fewer parameters, the M3 F models outperform state-ofthe-art latent factor approaches on all benchmarks, yielding the greatest gains in accuracy on sparsely-rated, high-variance items.
Lester W. Mackey, David Weiss, Michael I. Jordan
Added 09 Nov 2010
Updated 09 Nov 2010
Type Conference
Year 2010
Where ICML
Authors Lester W. Mackey, David Weiss, Michael I. Jordan
Comments (0)