Inference by belief propagation in composite systems

Abstract

We devise a message passing algorithm for probabilistic inference in composite systems, consisting of a large number of variables, that exhibit weak random interactions among all variables and strong interactions with a small subset of randomly chosen variables; the relative strength of the two interactions is controlled by a free parameter. We examine the performance of the algorithm numerically on a number of systems of this type for varying mixing parameter values.

Publication DOI: https://doi.org/10.1103/PhysRevE.78.021107
Divisions: College of Engineering & Physical Sciences > Systems analytics research institute (SARI)
Additional Information: ©2008 The American Physical Society
Uncontrolled Keywords: message passing algorithm,probabilistic inference,composite systems,variables,interactions,Mathematical Physics,Physics and Astronomy(all),Condensed Matter Physics,Statistical and Nonlinear Physics
Publication ISSN: 1550-2376
Last Modified: 18 Mar 2024 08:10
Date Deposited: 11 Mar 2019 17:50
Full Text Link: http://pre.aps. ... /v78/i2/e021107
Related URLs: http://www.scop ... tnerID=8YFLogxK (Scopus URL)
PURE Output Type: Article
Published Date: 2008-08-08
Authors: Mallard, Etienne
Saad, David (ORCID Profile 0000-0001-9821-2623)

Export / Share Citation


Statistics

Additional statistics for this record