On a Convergence Property of a Geometrical Algorithm for Statistical Manifolds

Shotaro Akaho*, Hideitsu Hino, Noboru Murata

*この研究の対応する著者

研究成果: Conference contribution

抄録

In this paper, we examine a geometrical projection algorithm for statistical inference. The algorithm is based on Pythagorean relation and it is derivative-free as well as representation-free that is useful in nonparametric cases. We derive a bound of learning rate to guarantee local convergence. In special cases of m-mixture and e-mixture estimation problems, we calculate specific forms of the bound that can be used easily in practice.

本文言語English
ホスト出版物のタイトルNeural Information Processing - 26th International Conference, ICONIP 2019, Proceedings
編集者Tom Gedeon, Kok Wai Wong, Minho Lee
出版社Springer
ページ262-272
ページ数11
ISBN(印刷版)9783030368012
DOI
出版ステータスPublished - 2019
イベント26th International Conference on Neural Information Processing, ICONIP 2019 - Sydney, Australia
継続期間: 2019 12 122019 12 15

出版物シリーズ

名前Communications in Computer and Information Science
1143 CCIS
ISSN(印刷版)1865-0929
ISSN(電子版)1865-0937

Conference

Conference26th International Conference on Neural Information Processing, ICONIP 2019
国/地域Australia
CitySydney
Period19/12/1219/12/15

ASJC Scopus subject areas

  • コンピュータ サイエンス(全般)
  • 数学 (全般)

フィンガープリント

「On a Convergence Property of a Geometrical Algorithm for Statistical Manifolds」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

引用スタイル