Non-parametric e-mixture of density functions

Hideitsu Hino; Ken Takano; Shotaro Akaho; Noboru Murata

doi:10.1007/978-3-319-46672-9_1

Non-parametric e-mixture of density functions

Hideitsu Hino^*, Ken Takano, Shotaro Akaho, Noboru Murata

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference 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 language	English
Title of host publication	Neural Information Processing - 23rd International Conference, ICONIP 2016, Proceedings
Publisher	Springer Verlag
Pages	3-10
Number of pages	8
Volume	9948 LNCS
ISBN (Print)	9783319466712
DOIs	https://doi.org/10.1007/978-3-319-46672-9_1
Publication status	Published - 2016
Event	23rd International Conference on Neural Information Processing, ICONIP 2016 - Kyoto, Japan Duration: 2016 Oct 16 → 2016 Oct 21

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	9948 LNCS
ISSN (Print)	03029743
ISSN (Electronic)	16113349

Other

Other	23rd International Conference on Neural Information Processing, ICONIP 2016
Country/Territory	Japan
City	Kyoto
Period	16/10/16 → 16/10/21

Keywords

Information geometry
Mixture model
Non-parametric method

ASJC Scopus subject areas

Theoretical Computer Science
Computer Science(all)

Access to Document

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

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 et al.

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 proceeding › Conference 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)). doi: 10.1007/978-3-319-46672-9_1

Hino, Hideitsu ; Takano, Ken ; Akaho, Shotaro et al. / 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",

note = "23rd International Conference on Neural Information Processing, ICONIP 2016 ; Conference date: 16-10-2016 Through 21-10-2016",

}

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

T2 - 23rd International Conference on Neural Information Processing, ICONIP 2016

Y2 - 16 October 2016 through 21 October 2016

ER -

Non-parametric e-mixture of density functions

Abstract

Publication series

Other

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this