A computationally efficient information estimator for weighted data

Hideitsu Hino, Noboru Murata

研究成果: Conference contribution

3 被引用数 (Scopus)

抄録

The Shannon information content is a fundamental quantity and it is of great importance to estimate it from observed dataset in the field of statistics, information theory, and machine learning. In this study, an estimator for the information content using a given set of weighted data is proposed. The empirical data distribution varies depending on the weight. The notable features of the proposed estimator are its computational efficiency and its ability to deal with weighted data. The proposed estimator is extended in order to estimate cross entropy, entropy and KL divergence with weighted data. Then, the estimators are applied to classification with one-class samples, and distribution preserving data compression problems.

本文言語English
ホスト出版物のタイトルArtificial Neural Networks and Machine Learning, ICANN 2011 - 21st International Conference on Artificial Neural Networks, Proceedings
ページ301-308
ページ数8
PART 2
DOI
出版ステータスPublished - 2011
イベント21st International Conference on Artificial Neural Networks, ICANN 2011 - Espoo, Finland
継続期間: 2011 6 142011 6 17

出版物シリーズ

名前Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
番号PART 2
6792 LNCS
ISSN(印刷版)0302-9743
ISSN(電子版)1611-3349

Conference

Conference21st International Conference on Artificial Neural Networks, ICANN 2011
CountryFinland
CityEspoo
Period11/6/1411/6/17

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

フィンガープリント 「A computationally efficient information estimator for weighted data」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