Direct Density Ratio Estimation for Large-scale Covariate Shift Adaptation

• Item type: JInfP(1)
• 公開日: 2009-04-08
• タイトル: Direct Density Ratio Estimation for Large-scale Covariate Shift Adaptation
• タイトル / 言語: en
• タイトル: Direct Density Ratio Estimation for Large-scale Covariate Shift Adaptation
• 言語: eng
• キーワード / 主題Scheme: Other
• キーワード / 主題: Regular Paper
• 資源タイプ / 資源タイプ識別子: http://purl.org/coar/resource_type/c_6501
• 資源タイプ: journal article
• その他タイトル / その他のタイトル: Knowledge Processing
• 著者所属: Tokyo Research Laboratory, IBM Research
• 著者所属: Tokyo Research Laboratory, IBM Research
• 著者所属: Tokyo Research Laboratory, IBM Research
• 著者所属: Department of Computer Science, University of Potsdam, Germany
• 著者所属: Department of Computer Science, Tokyo Institute of Technology
• 著者所属(英): en
• 著者所属(英): Tokyo Research Laboratory, IBM Research
• 著者所属(英): en
• 著者所属(英): Tokyo Research Laboratory, IBM Research
• 著者所属(英): en
• 著者所属(英): Tokyo Research Laboratory, IBM Research
• 著者所属(英): en
• 著者所属(英): Department of Computer Science, University of Potsdam, Germany
• 著者所属(英): en
• 著者所属(英): Department of Computer Science, Tokyo Institute of Technology
• 著者名: Yuta, Tsuboi; Hisashi, Kashima; Shohei, Hido; Steffen, Bickel; Masashi, Sugiyama
• 著者名(英): Yuta, Tsuboi; Hisashi, Kashima; Shohei, Hido; Steffen, Bickel; Masashi, Sugiyama
• 論文抄録 / 内容記述タイプ: Other
• 論文抄録 / 内容記述: <i>Covariate shift</i> is a situation in supervised learning where training and test <i>inputs</i> follow different distributions even though the functional relation remains unchanged. A common approach to compensating for the bias caused by covariate shift is to reweight the loss function according to the <i>importance</i>, which is the ratio of test and training densities. We propose a novel method that allows us to directly estimate the importance from samples without going through the hard task of density estimation. An advantage of the proposed method is that the computation time is nearly independent of the number of test input samples, which is highly beneficial in recent applications with large numbers of unlabeled samples. We demonstrate through experiments that the proposed method is computationally more efficient than existing approaches with comparable accuracy. We also describe a promising result for large-scale covariate shift adaptation in a natural language processing task.
• 論文抄録(英) / 内容記述タイプ: Other
• 論文抄録(英) / 内容記述: <i>Covariate shift</i> is a situation in supervised learning where training and test <i>inputs</i> follow different distributions even though the functional relation remains unchanged. A common approach to compensating for the bias caused by covariate shift is to reweight the loss function according to the <i>importance</i>, which is the ratio of test and training densities. We propose a novel method that allows us to directly estimate the importance from samples without going through the hard task of density estimation. An advantage of the proposed method is that the computation time is nearly independent of the number of test input samples, which is highly beneficial in recent applications with large numbers of unlabeled samples. We demonstrate through experiments that the proposed method is computationally more efficient than existing approaches with comparable accuracy. We also describe a promising result for large-scale covariate shift adaptation in a natural language processing task.
• 書誌レコードID / 収録物識別子タイプ: NCID
• 書誌レコードID / 収録物識別子: AA00700121
• 書誌情報: Journal of information processing; 巻 17, p. 138-155, 発行日 2009-04-08
• ISSN / 収録物識別子タイプ: ISSN
• ISSN / 収録物識別子: 1882-6652
• 出版者 / 言語: ja
• 出版者: 情報処理学会