Newton-like methods for nonparametric independent component analysis

Hao Shen*, Knut Hüper, Alexander J. Smola

*Corresponding author for this work

    Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

    4 Citations (Scopus)

    Abstract

    The performance of ICA algorithms significantly depends on the choice of the contrast function and the optimisation algorithm used in obtaining the demixing matrix. In this paper we focus on the standard linear nonparametric ICA problem from an optimisation point of view. It is well known that after a pre-whitening process, the problem can be solved via an optimisation approach on a suitable manifold. We propose an approximate Newton's method on the unit sphere to solve the one-unit linear nonparametric ICA problem. The local convergence properties are discussed. The performance of the proposed algorithms is investigated by numerical experiments.

    Original languageEnglish
    Title of host publicationNeural Information Processing - 13th International Conference, ICONIP 2006, Proceedings
    PublisherSpringer Verlag
    Pages1068-1077
    Number of pages10
    ISBN (Print)3540464794, 9783540464792
    DOIs
    Publication statusPublished - 2006
    Event13th International Conference on Neural Information Processing, ICONIP 2006 - Hong Kong, China
    Duration: 3 Oct 20066 Oct 2006

    Publication series

    NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    Volume4232 LNCS
    ISSN (Print)0302-9743
    ISSN (Electronic)1611-3349

    Conference

    Conference13th International Conference on Neural Information Processing, ICONIP 2006
    Country/TerritoryChina
    CityHong Kong
    Period3/10/066/10/06

    Fingerprint

    Dive into the research topics of 'Newton-like methods for nonparametric independent component analysis'. Together they form a unique fingerprint.

    Cite this