A quasi-likelihood approach to nonnegative matrix factorization

Karthik Devarajan, Vincent C.K. Cheung

Research output: Contribution to journalArticlepeer-review

6 Scopus citations


A unified approach to nonnegativematrix factorization based on the theory of generalized linear models is proposed. This approach embeds a variety of statistical models, including the exponential family, within a single theoretical framework and provides a unified view of such factorizations from the perspective of quasi-likelihood. Using this framework, a family of algorithms for handling signal-dependent noise is developed and its convergence proved using the expectation-maximization algorithm. In addition, ameasure to evaluate the goodness of fit of the resulting factorization is described. The proposed methods allowmodeling of nonlinear effects using appropriate link functions and are illustrated using an application in biomedical signal processing.

Original languageEnglish
Pages (from-to)1663-1693
Number of pages31
JournalNeural Computation
Issue number8
StatePublished - Aug 1 2016
Externally publishedYes


Dive into the research topics of 'A quasi-likelihood approach to nonnegative matrix factorization'. Together they form a unique fingerprint.

Cite this