Variational Bayes and Mean Field Approximations for Markov Field Unsupervised Estimation

Abstract : We consider the problem of parameter estimation of Markovian models where the exact computation of the partition function is not possible or computationally too expensive with MCMC methods. The main idea is then to approximate the expression of the likelihood by a simpler one where we can either have an analytical expression or compute it more efficiently. We consider two approaches: Variational Bayes Approximation (VBA) and Mean Field Approximation (MFA) and study the properties of such approximations and their effects on the estimation of the parameters.
Complete list of metadatas

https://hal.archives-ouvertes.fr/hal-00445310
Contributor : Hacheme Ayasso <>
Submitted on : Friday, January 8, 2010 - 10:54:33 AM
Last modification on : Thursday, April 5, 2018 - 12:30:04 PM

Links full text

Identifiers

Collections

Citation

Ali Mohammad-Djafari, Hacheme Ayasso. Variational Bayes and Mean Field Approximations for Markov Field Unsupervised Estimation. IEEE International Workshop on Machine Learning for Signal Processing., Sep 2009, Grenoble, France. pp.1-6, ⟨10.1109/MLSP.2009.5306261⟩. ⟨hal-00445310⟩

Share

Metrics

Record views

310