Presentation	2004/10/12 Improvement of Cluster BP Implementation Yusuke IZUMI, Taisuke SATO,
PDF Download Page	PDF download Page Link
Abstract(in Japanese)	(See Japanese page)
Abstract(in English)	Cluster BP, which has been proposed from a standpoint of statistical physics, is a method of approximate computation of marginal distributions of probablistic models including Bayesian networks. It is experimentally applied to decoding of error correcting code and linkage analysis in genetics. In this paper, we propose a method ofeffective computation of the values called message, so that the time of computation is reduced. We also attempt to improve the convergence of cluster BP by simple tricks.
Keyword(in Japanese)	(See Japanese page)
Keyword(in English)	Cluster BP / Junction Tree Algorithm / Convergence of BP
Paper #	NC2004-74
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)	Improvement of Cluster BP Implementation
Sub Title (in English)
Keyword(1)	Cluster BP
Keyword(2)	Junction Tree Algorithm
Keyword(3)	Convergence of BP
1st Author's Name	Yusuke IZUMI
1st Author's Affiliation	Graduate School of Infomation Science and Engineering, Toyko Institute of Technology()
2nd Author's Name	Taisuke SATO
2nd Author's Affiliation	Graduate School of Infomation Science and Engineering, Toyko Institute of Technology
Date	2004/10/12
Paper #	NC2004-74
Volume (vol)	vol.104
Number (no)	349
Page	pp.pp.-
#Pages	5
Date of Issue