Non-parametric e-mixture of density functions

Hideitsu Hino, Ken Takano, Shotaro Akaho, Noboru Murata

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

    1 Citation (Scopus)

    Abstract

    Mixture modeling is one of the simplest ways to represent complicated probability density functions, and to integrate information from different sources. There are two typical mixtures in the context of information geometry, the m-and e-mixtures. This paper proposes a novel framework of non-parametric e-mixture modeling by using a simple estimation algorithm based on geometrical insights into the characteristics of the e-mixture. An experimental result supports the proposed framework.

    Original languageEnglish
    Title of host publicationNeural Information Processing - 23rd International Conference, ICONIP 2016, Proceedings
    PublisherSpringer Verlag
    Pages3-10
    Number of pages8
    Volume9948 LNCS
    ISBN (Print)9783319466712
    DOIs
    Publication statusPublished - 2016
    Event23rd International Conference on Neural Information Processing, ICONIP 2016 - Kyoto, Japan
    Duration: 2016 Oct 162016 Oct 21

    Publication series

    NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    Volume9948 LNCS
    ISSN (Print)03029743
    ISSN (Electronic)16113349

    Other

    Other23rd International Conference on Neural Information Processing, ICONIP 2016
    CountryJapan
    CityKyoto
    Period16/10/1616/10/21

    Fingerprint

    Density Function
    Probability density function
    Mixture Modeling
    Information Geometry
    Estimation Algorithms
    Integrate
    Experimental Results
    Geometry
    Framework
    Context

    Keywords

    • Information geometry
    • Mixture model
    • Non-parametric method

    ASJC Scopus subject areas

    • Theoretical Computer Science
    • Computer Science(all)

    Cite this

    Hino, H., Takano, K., Akaho, S., & Murata, N. (2016). Non-parametric e-mixture of density functions. In Neural Information Processing - 23rd International Conference, ICONIP 2016, Proceedings (Vol. 9948 LNCS, pp. 3-10). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9948 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-319-46672-9_1

    Non-parametric e-mixture of density functions. / Hino, Hideitsu; Takano, Ken; Akaho, Shotaro; Murata, Noboru.

    Neural Information Processing - 23rd International Conference, ICONIP 2016, Proceedings. Vol. 9948 LNCS Springer Verlag, 2016. p. 3-10 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9948 LNCS).

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

    Hino, H, Takano, K, Akaho, S & Murata, N 2016, Non-parametric e-mixture of density functions. in Neural Information Processing - 23rd International Conference, ICONIP 2016, Proceedings. vol. 9948 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 9948 LNCS, Springer Verlag, pp. 3-10, 23rd International Conference on Neural Information Processing, ICONIP 2016, Kyoto, Japan, 16/10/16. https://doi.org/10.1007/978-3-319-46672-9_1
    Hino H, Takano K, Akaho S, Murata N. Non-parametric e-mixture of density functions. In Neural Information Processing - 23rd International Conference, ICONIP 2016, Proceedings. Vol. 9948 LNCS. Springer Verlag. 2016. p. 3-10. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-319-46672-9_1
    Hino, Hideitsu ; Takano, Ken ; Akaho, Shotaro ; Murata, Noboru. / Non-parametric e-mixture of density functions. Neural Information Processing - 23rd International Conference, ICONIP 2016, Proceedings. Vol. 9948 LNCS Springer Verlag, 2016. pp. 3-10 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
    @inproceedings{46fbca8b2a9d46018bec727b95a6e484,
    title = "Non-parametric e-mixture of density functions",
    abstract = "Mixture modeling is one of the simplest ways to represent complicated probability density functions, and to integrate information from different sources. There are two typical mixtures in the context of information geometry, the m-and e-mixtures. This paper proposes a novel framework of non-parametric e-mixture modeling by using a simple estimation algorithm based on geometrical insights into the characteristics of the e-mixture. An experimental result supports the proposed framework.",
    keywords = "Information geometry, Mixture model, Non-parametric method",
    author = "Hideitsu Hino and Ken Takano and Shotaro Akaho and Noboru Murata",
    year = "2016",
    doi = "10.1007/978-3-319-46672-9_1",
    language = "English",
    isbn = "9783319466712",
    volume = "9948 LNCS",
    series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
    publisher = "Springer Verlag",
    pages = "3--10",
    booktitle = "Neural Information Processing - 23rd International Conference, ICONIP 2016, Proceedings",
    address = "Germany",

    }

    TY - GEN

    T1 - Non-parametric e-mixture of density functions

    AU - Hino, Hideitsu

    AU - Takano, Ken

    AU - Akaho, Shotaro

    AU - Murata, Noboru

    PY - 2016

    Y1 - 2016

    N2 - Mixture modeling is one of the simplest ways to represent complicated probability density functions, and to integrate information from different sources. There are two typical mixtures in the context of information geometry, the m-and e-mixtures. This paper proposes a novel framework of non-parametric e-mixture modeling by using a simple estimation algorithm based on geometrical insights into the characteristics of the e-mixture. An experimental result supports the proposed framework.

    AB - Mixture modeling is one of the simplest ways to represent complicated probability density functions, and to integrate information from different sources. There are two typical mixtures in the context of information geometry, the m-and e-mixtures. This paper proposes a novel framework of non-parametric e-mixture modeling by using a simple estimation algorithm based on geometrical insights into the characteristics of the e-mixture. An experimental result supports the proposed framework.

    KW - Information geometry

    KW - Mixture model

    KW - Non-parametric method

    UR - http://www.scopus.com/inward/record.url?scp=84992666022&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84992666022&partnerID=8YFLogxK

    U2 - 10.1007/978-3-319-46672-9_1

    DO - 10.1007/978-3-319-46672-9_1

    M3 - Conference contribution

    AN - SCOPUS:84992666022

    SN - 9783319466712

    VL - 9948 LNCS

    T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

    SP - 3

    EP - 10

    BT - Neural Information Processing - 23rd International Conference, ICONIP 2016, Proceedings

    PB - Springer Verlag

    ER -