Proximal alternating-direction message-passing for MAP LP relaxation

Zhang, G; Heusdens, R

Proximal alternating-direction message-passing for MAP LP relaxation

Zhang, G

Heusdens, R

Permalink

Publication Type:: Conference Proceeding
Citation:: ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, 2013, pp. 3402 - 3406
Issue Date:: 2013-10-18

Closed Access

	Filename	Description	Size
	06638289.pdf	Published version	131.53 kB	Adobe PDF	View/Open

Copyright Clearance Process

Recently Added
In Progress
Closed Access

This item is closed access and not available.

Full metadata record

Field	Value	Language
dc.contributor.author	Zhang, G https://orcid.org/0000-0003-4521-542X	en_US
dc.contributor.author	Heusdens, R	en_US
dc.date.issued	2013-10-18	en_US
dc.identifier.citation	ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, 2013, pp. 3402 - 3406	en_US
dc.identifier.isbn	9781479903566	en_US
dc.identifier.issn	1520-6149	en_US
dc.identifier.uri	http://hdl.handle.net/10453/106084
dc.description.abstract	Linear programming (LP) relaxation for MAP inference over (factor) graphic models is one of the fundamental problems in machine learning. In this paper, we propose a new message-passing algorithm for the MAP LP-relaxation by using the proximal alternating-direction method of multipliers (PADMM). At each iteration, the new algorithm performs two layers of optimization, that is node-oriented optimization and factor-oriented optimization. On the other hand, the recently proposed augmented primal LP (APLP) algorithm, based on the ADMM, has to perform three layers of optimization. Our algorithm simplifies the APLP algorithm by removing one layer of optimization, thus reducing the computational complexities and further accelerating the convergence rate. We refer to our new algorithm as the proximal alternating-direction (PAD) algorithm. Experimental results confirm that the PAD algorithm indeed converges faster than the APLP method. © 2013 IEEE.	en_US
dc.relation.ispartof	ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings	en_US
dc.relation.isbasedon	10.1109/ICASSP.2013.6638289	en_US
dc.title	Proximal alternating-direction message-passing for MAP LP relaxation	en_US
dc.type	Conference Proceeding
utslib.for	080199 Artificial Intelligence and Image Processing not elsewhere classified	en_US
utslib.for	080599 Distributed Computing not elsewhere classified	en_US
utslib.for	090609 Signal Processing	en_US
pubs.embargo.period	Not known	en_US
pubs.organisational-group	/University of Technology Sydney
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology
utslib.copyright.status	closed_access
pubs.publication-status	Published	en_US

Abstract:

Linear programming (LP) relaxation for MAP inference over (factor) graphic models is one of the fundamental problems in machine learning. In this paper, we propose a new message-passing algorithm for the MAP LP-relaxation by using the proximal alternating-direction method of multipliers (PADMM). At each iteration, the new algorithm performs two layers of optimization, that is node-oriented optimization and factor-oriented optimization. On the other hand, the recently proposed augmented primal LP (APLP) algorithm, based on the ADMM, has to perform three layers of optimization. Our algorithm simplifies the APLP algorithm by removing one layer of optimization, thus reducing the computational complexities and further accelerating the convergence rate. We refer to our new algorithm as the proximal alternating-direction (PAD) algorithm. Experimental results confirm that the PAD algorithm indeed converges faster than the APLP method. © 2013 IEEE.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/106084