Change search
ReferencesLink to record
Permanent link

Direct link
Gaussian belief with dynamic data and in dynamic network
KTH, School of Electrical Engineering (EES), Centres, ACCESS Linnaeus Centre.
KTH, School of Electrical Engineering (EES), Centres, ACCESS Linnaeus Centre.
2009 (English)In: Europhysics letters, ISSN 0295-5075, E-ISSN 1286-4854, Vol. 87, no 6Article in journal (Refereed) Published
Abstract [en]

In this paper we analyze Belief Propagation over a Gaussian model in a dynamic environment. Recently, this has been proposed as a method to average local measurement values by a distributed protocol (Consensus Propagation, Moallemi C. C. and Van Roy B., IEEE Trans. Inf. Theory, 52 (2006) 4753) where the average is available for read-out at every single node. In the case that the underlying network is constant but the values to be averaged fluctuate ("dynamic data"), convergence and accuracy are determined by the spectral properties of an associated Ruelle-Perron-Frobenius operator. For Gaussian models on Erdos-Renyi graphs, numerical computation points to a spectral gap remaining in the large- size limit, implying exceptionally good scalability. In a model where the underlying network also fluctuates ("dynamic network"), averaging is more effective than in the dynamic data case. Altogether, this implies very good performance of these methods in very large systems, and opens a new field of statistical physics of large (and dynamic) information systems.

Place, publisher, year, edition, pages
2009. Vol. 87, no 6
Keyword [en]
graphical models, propagation
URN: urn:nbn:se:kth:diva-18855DOI: 10.1209/0295-5075/87/68004ISI: 000270659600030ScopusID: 2-s2.0-79051469011OAI: diva2:336902
QC 20100525Available from: 2010-08-05 Created: 2010-08-05 Last updated: 2010-12-13Bibliographically approved

Open Access in DiVA

No full text

Other links

Publisher's full textScopus

Search in DiVA

By author/editor
Aurell, ErikPfitzner, Rene
By organisation
ACCESS Linnaeus Centre
In the same journal
Europhysics letters

Search outside of DiVA

GoogleGoogle Scholar
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

Altmetric score

Total: 24 hits
ReferencesLink to record
Permanent link

Direct link