Higher order fused regularization for supervised learning with grouped parameters

Koh Takeuchi, Yoshinobu Kawahara, Tomoharu Iwata

研究成果: 書籍/レポート タイプへの寄稿会議への寄与

3 被引用数 (Scopus)

抄録

We often encounter situations in supervised learning where there exist possibly groups that consist of more than two parameters. For example, we might work on parameters that correspond to words expressing the same meaning, music pieces in the same genre, and books released in the same year. Based on such auxiliary information, we could suppose that parameters in a group have similar roles in a problem and similar values. In this paper, we propose the Higher Order Fused (HOF) regularization that can incorporate smoothness among parameters with group structures as prior knowledge in supervised learning. We define the HOF penalty as the Lovász extension of a submodular higher-order potential function, which encourages parameters in a group to take similar estimated values when used as a regularizer. Moreover, we develop an efficient network flow algorithm for calculating the proximity operator for the regularized problem. We investigate the empirical performance of the proposed algorithm by using synthetic and real-world data.

本文言語英語
ホスト出版物のタイトルMachine Learning and Knowledge Discovery in Databases - European Conference, ECML PKDD 2015, Proceedings
編集者Annalisa Appice, João Gama, Vitor Santos Costa, João Gama, Alípio Jorge, Annalisa Appice, Annalisa Appice, Vitor Santos Costa, Alípio Jorge, Annalisa Appice, Pedro Pereira Rodrigues, Pedro Pereira Rodrigues, João Gama, Vitor Santos Costa, Soares Soares, Pedro Pereira Rodrigues, Soares Soares, Soares Soares, João Gama, Soares Soares, Alípio Jorge, Alípio Jorge, Pedro Pereira Rodrigues, Vitor Santos Costa
出版社Springer Verlag
ページ577-593
ページ数17
ISBN(印刷版)9783319235271, 9783319235271, 9783319235271, 9783319235271
DOI
出版ステータス出版済み - 2015
外部発表はい
イベントEuropean Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, ECML PKDD 2015 - Porto, ポルトガル
継続期間: 9月 7 20159月 11 2015

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)
9284
ISSN(印刷版)0302-9743
ISSN(電子版)1611-3349

会議

会議European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, ECML PKDD 2015
国/地域ポルトガル
CityPorto
Period9/7/159/11/15

!!!All Science Journal Classification (ASJC) codes

  • 理論的コンピュータサイエンス
  • コンピュータ サイエンス(全般)

フィンガープリント

「Higher order fused regularization for supervised learning with grouped parameters」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