Lévy NMF : un modèle robuste de séparation de sources non-négatives

Translated title of the contribution: Lévy NMF: a robust model for nonnegative source separation

Paul Magron, Roland Badeau, Antoine Liutkus

    Research output: Chapter in Book/Report/Conference proceedingConference contributionProfessional


    In this paper, we address the problem of robust source separation of nonnegative data. We introduce the PαS distributions, which are a subclass of the stable distributions family, to model the nonnegative latent sources. Since those distributions are heavy-tailed, they are expected to be robust to outliers. Considering the Lévy distribution, the only PαS distribution whose density admits a closed form expression, we propose a mixture model called Lévy Nonnegative Matrix Factorization (Lévy NMF). The model is estimated in a maximum-likelihood sense. We also derive an estimator of the sources which extends the validity of the generalized Wiener filtering to the PαS case. Experiments on musical spectrograms and fluorescence spectra highlight the potential of the Lévy NMF model for decomposing nonnegative data.
    Translated title of the contributionLévy NMF: a robust model for nonnegative source separation
    Original languageFrench
    Title of host publicationActes du XXVIème Colloque GRETSI
    Publication statusPublished - 5 Sept 2017
    Publication typeD3 Professional conference proceedings
    EventGRETSI conference -
    Duration: 5 Sept 20178 Sept 2017


    ConferenceGRETSI conference


    Dive into the research topics of 'Lévy NMF: a robust model for nonnegative source separation'. Together they form a unique fingerprint.

    Cite this