Restricted Boltzmann machines for collaborative filtering

TLDR

Most collaborative filtering methods struggle with very large datasets. The paper demonstrates that Restricted Boltzmann Machines can model user/movie rating data. The authors develop efficient learning and inference procedures for RBMs and apply them to the 100‑million‑rating Netflix dataset. RBMs slightly outperform tuned SVD models, and a linear combination of multiple RBM and SVD predictions reduces error by more than 6% compared to Netflix's own system.

Abstract

Most of the existing approaches to collaborative filtering cannot handle very large data sets. In this paper we show how a class of two-layer undirected graphical models, called Restricted Boltzmann Machines (RBM's), can be used to model tabular data, such as user's ratings of movies. We present efficient learning and inference procedures for this class of models and demonstrate that RBM's can be successfully applied to the Netflix data set, containing over 100 million user/movie ratings. We also show that RBM's slightly outperform carefully-tuned SVD models. When the predictions of multiple RBM models and multiple SVD models are linearly combined, we achieve an error rate that is well over 6% better than the score of Netflix's own system.