Fixing max-product: Convergent message passing algorithms for MAP LP-relaxations

Amir Globerson, Tommi Jaakkola

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

We present a novel message passing algorithm for approximating the MAP problem in graphical models. The algorithm is similar in structure to max-product but unlike max-product it always converges, and can be proven to find the exactMAP solution in various settings. The algorithm is derived via block coordinate descent in a dual of the LP relaxation ofMAP, but does not require any tunable parameters such as step size or tree weights. We also describe a generalization of the method to cluster based potentials. The new method is tested on synthetic and real-world problems, and compares favorably with previous approaches.

Original languageEnglish
Title of host publicationAdvances in Neural Information Processing Systems 20 - Proceedings of the 2007 Conference
StatePublished - 2009
Externally publishedYes
Event21st Annual Conference on Neural Information Processing Systems, NIPS 2007 - Vancouver, BC, Canada
Duration: 3 Dec 20076 Dec 2007

Publication series

NameAdvances in Neural Information Processing Systems 20 - Proceedings of the 2007 Conference

Conference

Conference21st Annual Conference on Neural Information Processing Systems, NIPS 2007
Country/TerritoryCanada
CityVancouver, BC
Period3/12/076/12/07

Fingerprint

Dive into the research topics of 'Fixing max-product: Convergent message passing algorithms for MAP LP-relaxations'. Together they form a unique fingerprint.

Cite this