Maximum-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 discuss generalization error bounds for them. 1
Cited by 899 works.
Authors
References
- Learning the parts of objects by non-negative matrix factorization
- Unsupervised Learning by Probabilistic Latent Semantic Analysis
- Latent semantic models for collaborative filtering
- Weighted low-rank approximations
- A rank minimization heuristic with application to minimum order system approximation
- Modeling User Rating Profiles For Collaborative Filtering
- CSDP, A C library for semidefinite programming
- Learning with matrix factorizations
- Collaborative Filtering: A Machine Learning Perspective
- A Generalization of Principal Components Analysis to the Exponential Family