Approximate inference in credal networks by variational mean field methods

Jaime Shinsuke Ide; Fabio Gagliardi Cozman

Approximate inference in credal networks by variational mean field methods

Jaime Shinsuke Ide, Fabio Gagliardi Cozman

Resultado de la investigación: Contribución a una conferencia › Artículo de conferencia

7 Citas (Scopus)

Resumen

© 2005 Society for Imprecise Probability: Theories and Applications, SIPTA. All rights reserved. Graph-theoretical representations for sets of probability measures (credal networks) generally display high complexity, and approximate inference seems to be a natural solution for large networks. This paper introduces a variational approach to approximate inference in credal networks: we show how to formulate mean field approximations using naive (fully factorized) and structured (tree-like) schemes. We discuss the computational advantages of the variational approach, and present examples that illustrate the mechanics of the proposal.

Idioma original	Inglés estadounidense
Estado	Publicada - 1 ene 2005
Publicado de forma externa	Sí
Evento	4th International Symposium on Imprecise Probabilities and Their Applications, ISIPTA 2005 - Duración: 1 ene 2005 → …

Conferencia

Conferencia	4th International Symposium on Imprecise Probabilities and Their Applications, ISIPTA 2005
Período	1/01/05 → …

Otros archivos y enlaces

Huella Profundice en los temas de investigación de 'Approximate inference in credal networks by variational mean field methods'. En conjunto forman una huella única.

Citar esto

@conference{a54974c3d123421ebd500d7323ec2f19,

title = "Approximate inference in credal networks by variational mean field methods",

abstract = "{\textcopyright} 2005 Society for Imprecise Probability: Theories and Applications, SIPTA. All rights reserved. Graph-theoretical representations for sets of probability measures (credal networks) generally display high complexity, and approximate inference seems to be a natural solution for large networks. This paper introduces a variational approach to approximate inference in credal networks: we show how to formulate mean field approximations using naive (fully factorized) and structured (tree-like) schemes. We discuss the computational advantages of the variational approach, and present examples that illustrate the mechanics of the proposal.",

author = "Ide, {Jaime Shinsuke} and Cozman, {Fabio Gagliardi}",

year = "2005",

month = jan,

day = "1",

language = "American English",

note = "4th International Symposium on Imprecise Probabilities and Their Applications, ISIPTA 2005 ; Conference date: 01-01-2005",

}

TY - CONF

T1 - Approximate inference in credal networks by variational mean field methods

AU - Ide, Jaime Shinsuke

AU - Cozman, Fabio Gagliardi

PY - 2005/1/1

Y1 - 2005/1/1

N2 - © 2005 Society for Imprecise Probability: Theories and Applications, SIPTA. All rights reserved. Graph-theoretical representations for sets of probability measures (credal networks) generally display high complexity, and approximate inference seems to be a natural solution for large networks. This paper introduces a variational approach to approximate inference in credal networks: we show how to formulate mean field approximations using naive (fully factorized) and structured (tree-like) schemes. We discuss the computational advantages of the variational approach, and present examples that illustrate the mechanics of the proposal.

AB - © 2005 Society for Imprecise Probability: Theories and Applications, SIPTA. All rights reserved. Graph-theoretical representations for sets of probability measures (credal networks) generally display high complexity, and approximate inference seems to be a natural solution for large networks. This paper introduces a variational approach to approximate inference in credal networks: we show how to formulate mean field approximations using naive (fully factorized) and structured (tree-like) schemes. We discuss the computational advantages of the variational approach, and present examples that illustrate the mechanics of the proposal.

UR - https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=33751377668&origin=inward

UR - https://www.scopus.com/inward/citedby.uri?partnerID=HzOxMe3b&scp=33751377668&origin=inward

M3 - Conference Paper

T2 - 4th International Symposium on Imprecise Probabilities and Their Applications, ISIPTA 2005

Y2 - 1 January 2005

ER -