Presentation 2018-11-05
[Poster Presentation] Analysis of Empirical Bayes Estimation for Three Parameter Group Lasso
Tsukasa Yoshida, Kazuho Watanabe,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) As sparse estimation methods, Lasso and Group Lasso are often used for wide applications. It is necessary for successful applications of these regularization methods to determine the regularization parameter appropriately. Although the empirical Bayes approach provides an effective method to estimate the regularization parameter, its solution has yet to be fully investigated. In this study, we discuss the estimation of the regularization parameter using the empirical Bayes method in Group Lasso. We show the characteristics of the empirical Bayes estimator on a simplified linear regression model, whose design matrix is identity and the dimension of regression coefficients is three. We also discuss the property of the MAP estimator with its regularization parameter estimated by the empirical Bayes method.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Group Lasso / empirical Bayes / MAP estimation / Laplace prior / automatic relevance determination
Paper # IBISML2018-93
Date of Issue 2018-10-29 (IBISML)

Conference Information
Committee IBISML
Conference Date 2018/11/5(3days)
Place (in Japanese) (See Japanese page)
Place (in English) Hokkaido Citizens Activites Center (Kaderu 2.7)
Topics (in Japanese) (See Japanese page)
Topics (in English) Information-Based Induction Science Workshop (IBIS2018)
Chair Hisashi Kashima(Kyoto Univ.)
Vice Chair Masashi Sugiyama(Univ. of Tokyo) / Koji Tsuda(Univ. of Tokyo)
Secretary Masashi Sugiyama(Nagoya Inst. of Tech.) / Koji Tsuda(AIST)
Assistant Tomoharu Iwata(NTT) / Shigeyuki Oba(Kyoto Univ.)

Paper Information
Registration To Technical Committee on Infomation-Based Induction Sciences and Machine Learning
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) [Poster Presentation] Analysis of Empirical Bayes Estimation for Three Parameter Group Lasso
Sub Title (in English)
Keyword(1) Group Lasso
Keyword(2) empirical Bayes
Keyword(3) MAP estimation
Keyword(4) Laplace prior
Keyword(5) automatic relevance determination
1st Author's Name Tsukasa Yoshida
1st Author's Affiliation Toyohashi University of Technology(Toyohashi Tech)
2nd Author's Name Kazuho Watanabe
2nd Author's Affiliation Toyohashi University of Technology(Toyohashi Tech)
Date 2018-11-05
Paper # IBISML2018-93
Volume (vol) vol.118
Number (no) IBISML-284
Page pp.pp.367-372(IBISML),
#Pages 6
Date of Issue 2018-10-29 (IBISML)