Presentation | 2006-03-15 EM Algorithm for Bayesian Networks using Belief Propagation Yoichi MOTOMURA, Naoki SAKAI, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | Bayesian networks are promising tools for practical probabilistic information processing models using high speed computer and huge data. When we use Bayesian networks constructed from statistical data, missing data is processed by EM algorithm. In this research, EM algorithm for Bayesian networks is discussed. When E-step of EM algorithm, probabilistic reasoning is used on Bayesian networks. We can choose some different reasoning algorithm like (loopy) belief propagation, sampling methods and so on. In this paper, experimental result of EM algorithm using loopy BP and sampling method are shown. We see convergence, speed for some different size of networks. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Bayesian network / Statistical Learning / Probabilistic reasoning / EM algorithm / Belief propagation |
Paper # | NC2005-120 |
Date of Issue |
Conference Information | |
Committee | NC |
---|---|
Conference Date | 2006/3/8(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) | EM Algorithm for Bayesian Networks using Belief Propagation |
Sub Title (in English) | |
Keyword(1) | Bayesian network |
Keyword(2) | Statistical Learning |
Keyword(3) | Probabilistic reasoning |
Keyword(4) | EM algorithm |
Keyword(5) | Belief propagation |
1st Author's Name | Yoichi MOTOMURA |
1st Author's Affiliation | AIST, Digital Human Research Center() |
2nd Author's Name | Naoki SAKAI |
2nd Author's Affiliation | Mathematical Systems, Inc. |
Date | 2006-03-15 |
Paper # | NC2005-120 |
Volume (vol) | vol.105 |
Number (no) | 657 |
Page | pp.pp.- |
#Pages | 6 |
Date of Issue |