Timezone: »

Probabilistic Matrix Factorization
Russ Salakhutdinov · Andriy Mnih

Tue Dec 04 09:30 AM -- 09:50 AM (PST) @ None

Many existing approaches to collaborative filtering can neither handle very large datasets nor easily deal with users who have very few ratings. In this paper we present Probabilistic Matrix Factorization (PMF) which scales linearly with the number of observations and, more importantly, performs well on the large, sparse, and very imbalanced Netflix dataset. We further extend the PMF model to include an adaptive prior on the model parameters and show how the model capacity can be controlled automatically. Finally, we introduce a constrained version of the PMF model that is based on the assumption that users who have rated similar sets of movies are likely to have similar preferences. The resulting model is able to generalize considerably better for users with very few ratings. When the predictions of multiple PMF models are linearly combined with the RBM models recently introduced by Salakhutdinov et.al., we achieve an error rate of 0.8861, that is almost 7\% better than the score of Netflix's own system.

Author Information

Russ Salakhutdinov (Carnegie Mellon University)
Andriy Mnih (DeepMind)

More from the Same Authors