On affine non-negative matrix factorization

Hans Laurberg, Lars Kai Hansen

    Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

    521 Downloads (Orbit)

    Abstract

    We generalize the non-negative matrix factorization (NMF) generative model to incorporate an explicit offset. Multiplicative estimation algorithms are provided for the resulting sparse affine NMF model. We show that the affine model has improved uniqueness properties and leads to more accurate identification of mixing and sources.
    Original languageEnglish
    Title of host publicationICASSP 2007, Hawaii, USA
    PublisherIEEE
    Publication date2007
    Publication statusPublished - 2007
    Event2007 IEEE International Conference on Acoustics, Speech and Signal Processing - Honolulu, United States
    Duration: 15 Apr 200720 Apr 2007
    Conference number: 32

    Conference

    Conference2007 IEEE International Conference on Acoustics, Speech and Signal Processing
    Number32
    Country/TerritoryUnited States
    CityHonolulu
    Period15/04/200720/04/2007

    Fingerprint

    Dive into the research topics of 'On affine non-negative matrix factorization'. Together they form a unique fingerprint.

    Cite this