Large-Margin Matrix Factorization
Nathan Srebro, Jason D. M. Rennie, Tommi Jaakkola · 2004
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-definite program, and present generalization error bounds based on analyzing the Rademacher complexity of low-norm factorizations. 1