PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

A comparison of variational and Markov Chain Monte Carlo methods for inference in partially observed stochastic dynamic systems.
Yuan Shen, Cedric Archambeau, Dan Cornford, Manfred Opper, John Shawe-Taylor and Remi Barillec
Journal of Signal Processing Systems 2009. ISSN 1939-8018 (Print) 1939-8115 (Online)


In recent years we have developed a novel variational method for partially observed systems governed by stochastic differential equations. In this paper we provide a comparison of the Variational Gaussian Process Smoother with an exact solution computed using a Hybrid Monte Carlo approach to path sample sampling, applied to a stochastic double well potential model. It is demonstrated that the variational smoother provides us a very accurate estimate of mean path while conditional variance is slightly underestimated. We conclude with some remarks as to the advantages and disadvantages of the variational smoother.

EPrint Type:Article
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Computational, Information-Theoretic Learning with Statistics
Learning/Statistics & Optimisation
ID Code:5263
Deposited By:Manfred Opper
Deposited On:24 March 2009