Approximate Inference in Probabilistic Models

Manfred Opper, Ole Winther

AbstractAbstract
We present a framework for approximate inference in probabilistic data models which is based on free energies. The free energy is constructed from two approximating distributions which encode different aspects of the intractable model. Consistency between distributions is required on a chosen set of moments. We find good performance using sets of moments which either specify factorized nodes or a spanning tree on the nodes.
TypeConference paper [Without referee]
ConferenceAlgorithmic Learning Theory: 15th International Conference, ALT 2004, Padova, Italy, October 2-5, 2004. Proceedings
EditorsShai Ben-David, John Case, Akira Maruoka
Year2004    Month September    Vol. 3244    pp. 494 - 504
PublisherSpringer-Verlag Heidelberg
SeriesLecture Notes in Computer Science
ISBN / ISSN3-540-23356-3
Electronic version(s)[pdf]
Publication linkhttp://www.springerlink.com/app/home/contribution.asp?wasp=4h83eykqtlckpl98rlrl&referrer=parent&backto=issue,37,37;journal,80,1851;linkingpublicationresults,1:105633,1
BibTeX data [bibtex]
IMM Group(s)Intelligent Signal Processing