Title | ||
---|---|---|
A new Bayesian approach to nonnegative matrix factorization: Uniqueness and model order selection. |
Abstract | ||
---|---|---|
NMF is a blind source separation technique decomposing multivariate non-negative data sets into meaningful non-negative basis components and non-negative weights. There are still open problems to be solved: uniqueness and model order selection as well as developing efficient NMF algorithms for large scale problems. Addressing uniqueness issues, we propose a Bayesian optimality criterion (BOC) for NMF solutions which can be derived in the absence of prior knowledge. Furthermore, we present a new Variational Bayes NMF algorithm VBNMF which is a straight forward generalization of the canonical Lee–Seung method for the Euclidean NMF problem and demonstrate its ability to automatically detect the actual number of components in non-negative data. |
Year | DOI | Venue |
---|---|---|
2014 | 10.1016/j.neucom.2014.02.021 | Neurocomputing |
Keywords | Field | DocType |
Bayes NMF,Variational Bayes,Bayesian optimality criterion,Generalized Lee–Seung update rules | Uniqueness,Data set,Optimality criterion,Pattern recognition,Non-negative matrix factorization,Artificial intelligence,Euclidean geometry,Blind signal separation,Machine learning,Mathematics,Bayes' theorem,Bayesian probability | Journal |
Volume | ISSN | Citations |
138 | 0925-2312 | 1 |
PageRank | References | Authors |
0.36 | 17 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Reinhard Schachtner | 1 | 17 | 3.10 |
Gerhard Pöppel | 2 | 17 | 3.44 |
Ana Maria Tomé | 3 | 163 | 30.42 |
Carlos García Puntonet | 4 | 107 | 25.86 |
Elmar Wolfgang Lang | 5 | 260 | 36.10 |