Presentation 2012-03-13
Feature Selection via l_1-Penalized Squared-Loss Mutual Information
Wittawat JITKRITTUM, Hirotaka HACHIYA, Masashi SUGIYAMA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Feature selection is a technique to screen out less important features. Many existing supervised feature selection algorithms use redundancy and relevancy as the main criteria to select features. However, feature interaction, potentially a key characteristic in real-world problems, has not received much attention. As an attempt to take feature interaction into account, we propose l_1-LSMI, an l_1-regularization based algorithm that maximizes a squared-loss variant of mutual information between selected features and outputs. Numerical results show that l_1-LSMI performs well in handling redundancy, detecting non-linear dependency, and considering feature interaction.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) feature selection / l_1-regularization / squared-loss mutual information / density-ratio estimation / dimensionality reduction
Paper # IBISML2011-107
Date of Issue

Conference Information
Committee IBISML
Conference Date 2012/3/5(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Information-Based Induction Sciences and Machine Learning (IBISML)
Language ENG
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Feature Selection via l_1-Penalized Squared-Loss Mutual Information
Sub Title (in English)
Keyword(1) feature selection
Keyword(2) l_1-regularization
Keyword(3) squared-loss mutual information
Keyword(4) density-ratio estimation
Keyword(5) dimensionality reduction
1st Author's Name Wittawat JITKRITTUM
1st Author's Affiliation Department of Computer Science, Tokyo Institute of Technology()
2nd Author's Name Hirotaka HACHIYA
2nd Author's Affiliation Department of Computer Science, Tokyo Institute of Technology
3rd Author's Name Masashi SUGIYAMA
3rd Author's Affiliation Department of Computer Science, Tokyo Institute of Technology
Date 2012-03-13
Paper # IBISML2011-107
Volume (vol) vol.111
Number (no) 480
Page pp.pp.-
#Pages 8
Date of Issue