Factor graph fragmentization of expectation propagation

Chen, WY; Wand, MP

Factor graph fragmentization of expectation propagation

Chen, WY Wand, MP

Permalink

Publisher:: SPRINGER HEIDELBERG
Publication Type:: Journal Article
Citation:: Journal of the Korean Statistical Society, 2018, 49, (3), pp. 722-756
Issue Date:: 2018

Closed Access

	Filename	Description	Size
	Chen-Wand2020_Article_FactorGraphFragmentizationOfEx.pdf	Published version	1.19 MB	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	Chen, WY
dc.contributor.author	Wand, MP
dc.date.accessioned	2021-03-08T07:40:41Z
dc.date.available	2021-03-08T07:40:41Z
dc.date.issued	2018
dc.identifier.citation	Journal of the Korean Statistical Society, 2018, 49, (3), pp. 722-756
dc.identifier.issn	1226-3192
dc.identifier.issn	2005-2863
dc.identifier.uri	http://hdl.handle.net/10453/146925
dc.description.abstract	Expectation propagation is a general approach to fast approximate inference for graphical models. The existing literature treats models separately when it comes to deriving and coding expectation propagation inference algorithms. This comes at the cost of similar, long-winded algebraic steps being repeated and slowing down algorithmic development. We demonstrate how factor graph fragmentization can overcome this impediment. This involves adoption of the message passing on a factor graph approach to expectation propagation and identification of factor graph sub-graphs, which we call fragments, that are common to wide classes of models. Key fragments and their corresponding messages are catalogued which means that their algebra does not need to be repeated. This allows compartmentalization of coding and efficient software development.
dc.language	English
dc.publisher	SPRINGER HEIDELBERG
dc.relation	http://purl.org/au-research/grants/arc/DP180100597
dc.relation.ispartof	Journal of the Korean Statistical Society
dc.relation.isbasedon	10.1007/s42952-019-00033-9
dc.rights	info:eu-repo/semantics/closedAccess
dc.subject	0104 Statistics
dc.title	Factor graph fragmentization of expectation propagation
dc.type	Journal Article
utslib.citation.volume	49
utslib.for	0104 Statistics
pubs.organisational-group	/University of Technology Sydney/Faculty of Science
pubs.organisational-group	/University of Technology Sydney/Faculty of Science/School of Mathematical and Physical Sciences
pubs.organisational-group	/University of Technology Sydney
utslib.copyright.status	closed_access	*
dc.date.updated	2021-03-08T07:40:40Z
pubs.issue	3
pubs.publication-status	Published
pubs.volume	49
utslib.citation.issue	3

Abstract:

Expectation propagation is a general approach to fast approximate inference for graphical models. The existing literature treats models separately when it comes to deriving and coding expectation propagation inference algorithms. This comes at the cost of similar, long-winded algebraic steps being repeated and slowing down algorithmic development. We demonstrate how factor graph fragmentization can overcome this impediment. This involves adoption of the message passing on a factor graph approach to expectation propagation and identification of factor graph sub-graphs, which we call fragments, that are common to wide classes of models. Key fragments and their corresponding messages are catalogued which means that their algebra does not need to be repeated. This allows compartmentalization of coding and efficient software development.

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

http://hdl.handle.net/10453/146925