Guided Selection of Accurate Belief Propagation Fixed Points

Research output: Contribution to conferencePaperResearchpeer-review

Abstract

Belief propagation (BP) and the Bethe approximation are two closely relatedconcepts that both suffer from the existence of multiple fixed points (or stationarypoints). We propose a modification of BP, termed self-guided belief propagation(SBP), that incorporates the pairwise potentials only gradually; this essentiallyselects one specific fixed point and increases the accuracy without increasing thecomputational burden. We apply SBP to various models with Ising potentials andshow that: (i) SBP is superior in terms of accuracy whenever BP converges, and (ii)SBP obtains a unique, stable, and accurate solution whenever BP does not converge.
Original languageEnglish
Number of pages6
Publication statusPublished - 14 Dec 2019
EventMachine Learning and the Physical Sciences - Vancouver, Canada
Duration: 14 Dec 201914 Dec 2019

Conference

ConferenceMachine Learning and the Physical Sciences
CountryCanada
CityVancouver
Period14/12/1914/12/19

Cite this

Knoll, C., Kulmer, F., & Pernkopf, F. (2019). Guided Selection of Accurate Belief Propagation Fixed Points. Paper presented at Machine Learning and the Physical Sciences, Vancouver, Canada.