Presentation	2010-06-15 Statistical Machine Learning Based on Nonparametric Bayesian Models Takaki MAKINO,
PDF Download Page	PDF download Page Link
Abstract(in Japanese)	(See Japanese page)
Abstract(in English)	Nonparametric Bayesian models are a new approach for machine learning, involving overfitting avoidance and model selection. Nonparametric Bayesian approach uses the technique of Bayesian inference, which avoids too complex solutions by introducing a prior probability on the hypothetical space, to the selection problem from an infinite number of models, and used in many applications, including function regression, clustering, and document topic models. In this review, we introduce infinite hidden Markov model, which is a nonparametric Bayesian version of hidden Markov model that can handle a possibly infinite number of hidden states, and its extension to hierarchical clustering of states.
Keyword(in Japanese)	(See Japanese page)
Keyword(in English)	nonparametric Bayesian models / hidden Markov models / Dirichlet process / clustering
Paper #	IBISML2010-14
Date of Issue

Conference Information
Committee	IBISML
Conference Date	2010/6/7(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	JPN
Title (in Japanese)	(See Japanese page)
Sub Title (in Japanese)	(See Japanese page)
Title (in English)	Statistical Machine Learning Based on Nonparametric Bayesian Models
Sub Title (in English)
Keyword(1)	nonparametric Bayesian models
Keyword(2)	hidden Markov models
Keyword(3)	Dirichlet process
Keyword(4)	clustering
1st Author's Name	Takaki MAKINO
1st Author's Affiliation	Division of Project Coordination, University of Tokyo()
Date	2010-06-15
Paper #	IBISML2010-14
Volume (vol)	vol.110
Number (no)	76
Page	pp.pp.-
#Pages	8
Date of Issue