Unitary Approximate Message Passing for Matrix Factorization
This provides an efficient solution for various matrix factorization problems in areas like compressive sensing and robust PCA, though it appears incremental as it builds on existing methods.
The paper tackled matrix factorization with constraints by developing a Bayesian approach using variational inference and unitary approximate message passing, resulting in UAMPMF, which significantly outperforms state-of-the-art algorithms in recovery accuracy, robustness, and computational complexity.
We consider matrix factorization (MF) with certain constraints, which finds wide applications in various areas. Leveraging variational inference (VI) and unitary approximate message passing (UAMP), we develop a Bayesian approach to MF with an efficient message passing implementation, called UAMPMF. With proper priors imposed on the factor matrices, UAMPMF can be used to solve many problems that can be formulated as MF, such as non negative matrix factorization, dictionary learning, compressive sensing with matrix uncertainty, robust principal component analysis, and sparse matrix factorization. Extensive numerical examples are provided to show that UAMPMF significantly outperforms state-of-the-art algorithms in terms of recovery accuracy, robustness and computational complexity.