Simultaneous safe screening of features and samples in doubly sparse modeling

Atsushi Shibagaki, Masayuki Karasuyama, Kohei Hatano, Ichiro Takeuchi

研究成果: 書籍/レポート タイプへの寄稿会議への寄与

18 被引用数 (Scopus)


The problem of learning a sparse model is conceptually interpreted as the process of identify-ing active features/samples and then optimizing the model over them. Recently introduced safe screening allows us to identify a part of non- Active features/samples. So far, safe screening has been individually studied either for feature screening or for sample screening. In this paper, we introduce a new approach for safely screening features and samples simultaneously by al-ternatively iterating feature and sample screening steps. A significant advantage of considering them simultaneously rather than individually is that they have a synergy effect in the sense that the results of the previous safe feature screening can be exploited for improving the next safe sample screening performances, and vice- versa. We first theoretically investigate the synergy effect, and then illustrate the practical advantage through intensive numerical experiments for problems with large numbers of features and samples.

ホスト出版物のタイトル33rd International Conference on Machine Learning, ICML 2016
編集者Kilian Q. Weinberger, Maria Florina Balcan
出版社International Machine Learning Society (IMLS)
出版ステータス出版済み - 2016
イベント33rd International Conference on Machine Learning, ICML 2016 - New York City, 米国
継続期間: 6月 19 20166月 24 2016


名前33rd International Conference on Machine Learning, ICML 2016


その他33rd International Conference on Machine Learning, ICML 2016
CityNew York City

!!!All Science Journal Classification (ASJC) codes

  • 人工知能
  • ソフトウェア
  • コンピュータ ネットワークおよび通信