Presentation | 2002/1/22 Graphical Models and Variational Principles for Mean Field Approximations Yoshiyuki KABASHIMA, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | Mean field approximations are practical methods to make large scale statistical models computationally tractable. In this paper, we show that a variation of mean field approximations, termed the Bethe approximation or the loopy belief propagation, can be derived by extending a variational principle which is validated for a class of statistical models characterized by junction trees to general graphical models. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Mean field approximations / Bethe approximation / belief propagation / junction trees / graphical models |
Paper # | |
Date of Issue |
Conference Information | |
Committee | NC |
---|---|
Conference Date | 2002/1/22(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) | Graphical Models and Variational Principles for Mean Field Approximations |
Sub Title (in English) | |
Keyword(1) | Mean field approximations |
Keyword(2) | Bethe approximation |
Keyword(3) | belief propagation |
Keyword(4) | junction trees |
Keyword(5) | graphical models |
1st Author's Name | Yoshiyuki KABASHIMA |
1st Author's Affiliation | Department of Computational Intelligence and Systems Science, Tokyo Institute of Technology() |
Date | 2002/1/22 |
Paper # | |
Volume (vol) | vol.101 |
Number (no) | 616 |
Page | pp.pp.- |
#Pages | 8 |
Date of Issue |