Approximate inference in credal networks by variational mean field methods

Jaime Shinsuke Ide, Fabio Gagliardi Cozman

Resultado de la investigación: Contribución a una conferenciaArtículo 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 originalInglés estadounidense
EstadoPublicada - 1 ene 2005
Publicado de forma externa
Evento4th International Symposium on Imprecise Probabilities and Their Applications, ISIPTA 2005 -
Duración: 1 ene 2005 → …

Conferencia

Conferencia4th International Symposium on Imprecise Probabilities and Their Applications, ISIPTA 2005
Período1/01/05 → …

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

    Ide, J. S., & Cozman, F. G. (2005). Approximate inference in credal networks by variational mean field methods. Papel presentado en 4th International Symposium on Imprecise Probabilities and Their Applications, ISIPTA 2005, .