Maximum-Margin Matrix Factorization

Part of Advances in Neural Information Processing Systems 17 (NIPS 2004)

Bibtex Metadata Paper


Nathan Srebro, Jason Rennie, Tommi Jaakkola


We present a novel approach to collaborative prediction, using low-norm instead of low-rank factorizations. The approach is inspired by, and has strong connections to, large-margin linear discrimination. We show how to learn low-norm factorizations by solving a semi-deļ¬nite program, and discuss generalization error bounds for them.