File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: Fixed Points of Gaussian Belief Propagation and Relation to Convergence

TitleFixed Points of Gaussian Belief Propagation and Relation to Convergence
Authors
KeywordsConvergence
Covariance matrices
Belief propagation
Mathematical model
Graphical models
Issue Date2019
PublisherIEEE. The Journal's web site is located at http://ieeexplore.ieee.org/xpl/RecentIssue.jsp?punumber=78
Citation
IEEE Transactions on Signal Processing, 2019, v. 67 n. 23, p. 6025-6038 How to Cite?
AbstractIn general, Gaussian belief propagation (BP) is not guaranteed to converge in loopy graphs. But when Gaussian BP converges to a fixed point of its update equations, the exact means of the marginal distributions can be obtained. However, given a Gaussian graphical model, whether Gaussian BP would have fixed points is unknown. Moreover, the relation between the convergence of Gaussian BP and its fixed points is not clear. To answer these questions, the necessary and sufficient existence condition and an easily verifiable sufficient existence condition of fixed points of Gaussian BP are proposed. Conditioned on the existence of fixed points of Gaussian BP, the convergence conditions of outgoing messages' parameters are analyzed, where outgoing message denotes the message flowing from a variable node to a factor node on a factor graph. It is proved that outgoing messages' precisions (the reciprocal of variance) would converge if there exist fixed points for Gaussian BP. This provides an elegant interpretation of the convergence condition of outgoing messages' precisions. On the other hand, to guarantee the convergence of outgoing messages' means, a sufficient condition for Gaussian models with a single loop and a sufficient condition that can be checked without obtaining the converged outgoing messages' precisions are derived. The relations between the convergence conditions of outgoing messages' parameters and those of incoming messages' parameters are also revealed, making any convergence condition of outgoing messages valid for verifying the convergence of incoming messages. Numerical results are presented to corroborate the newly established theories.
Persistent Identifierhttp://hdl.handle.net/10722/290168
ISSN
2020 Impact Factor: 4.931
2015 SCImago Journal Rankings: 2.004
ISI Accession Number ID

 

DC FieldValueLanguage
dc.contributor.authorLI, B-
dc.contributor.authorSu, Q-
dc.contributor.authorWu, YC-
dc.date.accessioned2020-10-22T08:23:00Z-
dc.date.available2020-10-22T08:23:00Z-
dc.date.issued2019-
dc.identifier.citationIEEE Transactions on Signal Processing, 2019, v. 67 n. 23, p. 6025-6038-
dc.identifier.issn1053-587X-
dc.identifier.urihttp://hdl.handle.net/10722/290168-
dc.description.abstractIn general, Gaussian belief propagation (BP) is not guaranteed to converge in loopy graphs. But when Gaussian BP converges to a fixed point of its update equations, the exact means of the marginal distributions can be obtained. However, given a Gaussian graphical model, whether Gaussian BP would have fixed points is unknown. Moreover, the relation between the convergence of Gaussian BP and its fixed points is not clear. To answer these questions, the necessary and sufficient existence condition and an easily verifiable sufficient existence condition of fixed points of Gaussian BP are proposed. Conditioned on the existence of fixed points of Gaussian BP, the convergence conditions of outgoing messages' parameters are analyzed, where outgoing message denotes the message flowing from a variable node to a factor node on a factor graph. It is proved that outgoing messages' precisions (the reciprocal of variance) would converge if there exist fixed points for Gaussian BP. This provides an elegant interpretation of the convergence condition of outgoing messages' precisions. On the other hand, to guarantee the convergence of outgoing messages' means, a sufficient condition for Gaussian models with a single loop and a sufficient condition that can be checked without obtaining the converged outgoing messages' precisions are derived. The relations between the convergence conditions of outgoing messages' parameters and those of incoming messages' parameters are also revealed, making any convergence condition of outgoing messages valid for verifying the convergence of incoming messages. Numerical results are presented to corroborate the newly established theories.-
dc.languageeng-
dc.publisherIEEE. The Journal's web site is located at http://ieeexplore.ieee.org/xpl/RecentIssue.jsp?punumber=78-
dc.relation.ispartofIEEE Transactions on Signal Processing-
dc.rightsIEEE Transactions on Signal Processing. Copyright © IEEE.-
dc.rights©20xx 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.-
dc.subjectConvergence-
dc.subjectCovariance matrices-
dc.subjectBelief propagation-
dc.subjectMathematical model-
dc.subjectGraphical models-
dc.titleFixed Points of Gaussian Belief Propagation and Relation to Convergence-
dc.typeArticle-
dc.identifier.emailWu, YC: ycwu@eee.hku.hk-
dc.identifier.authorityWu, YC=rp00195-
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.doi10.1109/TSP.2019.2951225-
dc.identifier.scopuseid_2-s2.0-85077809817-
dc.identifier.hkuros316735-
dc.identifier.volume67-
dc.identifier.issue23-
dc.identifier.spage6025-
dc.identifier.epage6038-
dc.identifier.isiWOS:000502125700001-
dc.publisher.placeUnited States-
dc.identifier.issnl1053-587X-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats