Active collaborative prediction with maximum margin matrix factorization

Abstract

Collaborative prediction (CP) is a problem of predicting unobserved entries in sparsely observed matrices, e.g. product ratings by different users in online recommender systems. However, the quality of prediction may be quite sensitive to the choice of available samples, which motivates active sampling approaches. In this paper, we suggest an active sampling method based on the recently proposed Maximum-Margin Matrix Factorization (MMMF) (Srebro, Rennie, & Jaakkola 2004), a linear factor model that was shown to outperform state-of-art collaborative prediction techniques. MMMF is formulated as a semi-definite program (SDP) that finds a low-norm (rather than traditional low-rank) matrix factorization, and is also closely related to learning max-margin linear discriminants (SVMs). This relation to SVMs inspires several margin-based active sampling heuristics that augment MMMF and demonstrate promising results in a variety of practical domains, including both traditional recommender systems and novel systems-management applications such as predicting latency and bandwidth in computer networks. Copyright © 2007, authors listed above. All rights reserved.