Evaluation of variational and Markov Chain Monte Carlo methods for inference in partially observed stochastic dynamic systems


In recent work we have developed a novel variational inference 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 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 marginal variance is slightly underestimated. We conclude with some remarks as to the advantages and disadvantages of the variational smoother.

Publication DOI: https://doi.org/10.1109/MLSP.2007.4414324
Divisions: College of Engineering & Physical Sciences > School of Informatics and Digital Engineering > Computer Science
College of Engineering & Physical Sciences > Systems analytics research institute (SARI)
Additional Information: © 2007 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
Event Title: 17th IEEE International Workshop on Machine Learning for Signal Processing, MLSP-2007
Event Type: Other
Event Dates: 2007-08-27 - 2007-08-29
Uncontrolled Keywords: Computer Science(all),Signal Processing
ISBN: 1424415667, 9781424415663
Full Text Link:
Related URLs: http://www.scop ... tnerID=8YFLogxK (Scopus URL)
https://ieeexpl ... ocument/4414324 (Publisher URL)
PURE Output Type: Conference contribution
Published Date: 2007-12-01
Authors: Shen, Y.
Archambeau, C.
Cornford, D. (ORCID Profile 0000-0001-8787-6758)
Opper, M.
Shawe-Taylor, J.
Barillec, R.



Version: Accepted Version

| Preview

Export / Share Citation


Additional statistics for this record