Presentation 2004/10/12
Geometrical Structure of Learning Algorithms
Noboru MURATA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In the machine learning, it is quite important how to model target problems, and also how to construct efficient algorithms in order to obtain optimal solutions. The concept of information geometry, in which geometrical structures of stochastic models are considered based on differential geometry, helps us to understand the mechanism of learning algorithms. In this article, geometrical understandings of learning algorithms are discussed, such as EM algorithm, boosting, bagging.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) information geometry / EM algorithm / boosting / bagging
Paper # NC2004-77
Date of Issue

Conference Information
Committee NC
Conference Date 2004/10/12(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 Neurocomputing (NC)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Geometrical Structure of Learning Algorithms
Sub Title (in English)
Keyword(1) information geometry
Keyword(2) EM algorithm
Keyword(3) boosting
Keyword(4) bagging
1st Author's Name Noboru MURATA
1st Author's Affiliation Waseda University()
Date 2004/10/12
Paper # NC2004-77
Volume (vol) vol.104
Number (no) 349
Page pp.pp.-
#Pages 6
Date of Issue