Parallel probabilistic inference by weighted model counting

Giso H. Dal; Alfons W. Laarman; Peter J.F. Lucas

Parallel probabilistic inference by weighted model counting

Giso H. Dal, Alfons W. Laarman, Peter J.F. Lucas

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

2 Citations (Scopus)

11 Downloads (Pure)

Abstract

Knowledge compilation as part of the Weighted Model Counting approach has proven to be an efficient tool for exact inference in probabilistic graphical models, by exploiting structures that more traditional methods can not. The availability of affordable high performance commodity hardware has been an inspiration for other inference approaches to exploit parallelism, to great suc- cess. In this paper, we explore the possibilities for Weighted Model Counting. We have empirically confirmed that exploited parallelism yields substantial speedups using a set of real-world Bayesian networks.

Original language	English
Title of host publication	Proceedings of the Ninth International Conference on Probabilistic Graphical Models
Subtitle of host publication	11-14 September 2018, Prague, Czech Republic
Editors	Václav Kratochvíl, Milan Studený
Publisher	MLResearchPress
Pages	97-108
Number of pages	12
Publication status	Published - 2018
Externally published	Yes
Event	9th International Conference on Probabilistic Graphical Models, PGM 2018 - Prague, Czech Republic Duration: 11 Sept 2018 → 14 Sept 2018 Conference number: 9

Publication series

Name	Proceedings of Machine Learning Research (PMLR)
Publisher	JMLR
Volume	72
ISSN (Electronic)	2640-3498

Conference

Conference	9th International Conference on Probabilistic Graphical Models, PGM 2018
Abbreviated title	PGM 2018
Country/Territory	Czech Republic
City	Prague
Period	11/09/18 → 14/09/18

Access to Document

ArticleFinal published version, 305 KB

https://proceedings.mlr.press/v72/dal18a.html

Cite this

Dal, G. H., Laarman, A. W., & Lucas, P. J. F. (2018). Parallel probabilistic inference by weighted model counting. In V. Kratochvíl, & M. Studený (Eds.), Proceedings of the Ninth International Conference on Probabilistic Graphical Models: 11-14 September 2018, Prague, Czech Republic (pp. 97-108). (Proceedings of Machine Learning Research (PMLR); Vol. 72). MLResearchPress. https://proceedings.mlr.press/v72/dal18a.html

Dal, Giso H. ; Laarman, Alfons W. ; Lucas, Peter J.F. / Parallel probabilistic inference by weighted model counting. Proceedings of the Ninth International Conference on Probabilistic Graphical Models: 11-14 September 2018, Prague, Czech Republic. editor / Václav Kratochvíl ; Milan Studený. MLResearchPress, 2018. pp. 97-108 (Proceedings of Machine Learning Research (PMLR)).

@inproceedings{6176f3967f0d45ac96455dc8e004941f,

title = "Parallel probabilistic inference by weighted model counting",

abstract = "Knowledge compilation as part of the Weighted Model Counting approach has proven to be an efficient tool for exact inference in probabilistic graphical models, by exploiting structures that more traditional methods can not. The availability of affordable high performance commodity hardware has been an inspiration for other inference approaches to exploit parallelism, to great suc- cess. In this paper, we explore the possibilities for Weighted Model Counting. We have empirically confirmed that exploited parallelism yields substantial speedups using a set of real-world Bayesian networks.",

author = "Dal, {Giso H.} and Laarman, {Alfons W.} and Lucas, {Peter J.F.}",

year = "2018",

language = "English",

series = "Proceedings of Machine Learning Research (PMLR)",

publisher = "MLResearchPress",

pages = "97--108",

editor = "V{\'a}clav Kratochv{\'i}l and Milan Studen{\'y}",

booktitle = "Proceedings of the Ninth International Conference on Probabilistic Graphical Models",

note = "9th International Conference on Probabilistic Graphical Models, PGM 2018, PGM 2018 ; Conference date: 11-09-2018 Through 14-09-2018",

}

Dal, GH, Laarman, AW & Lucas, PJF 2018, Parallel probabilistic inference by weighted model counting. in V Kratochvíl & M Studený (eds), Proceedings of the Ninth International Conference on Probabilistic Graphical Models: 11-14 September 2018, Prague, Czech Republic. Proceedings of Machine Learning Research (PMLR), vol. 72, MLResearchPress, pp. 97-108, 9th International Conference on Probabilistic Graphical Models, PGM 2018, Prague, Czech Republic, 11/09/18. <https://proceedings.mlr.press/v72/dal18a.html>

Parallel probabilistic inference by weighted model counting. / Dal, Giso H.; Laarman, Alfons W.; Lucas, Peter J.F.
Proceedings of the Ninth International Conference on Probabilistic Graphical Models: 11-14 September 2018, Prague, Czech Republic. ed. / Václav Kratochvíl; Milan Studený. MLResearchPress, 2018. p. 97-108 (Proceedings of Machine Learning Research (PMLR); Vol. 72).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Parallel probabilistic inference by weighted model counting

AU - Dal, Giso H.

AU - Laarman, Alfons W.

AU - Lucas, Peter J.F.

N1 - Conference code: 9

PY - 2018

Y1 - 2018

N2 - Knowledge compilation as part of the Weighted Model Counting approach has proven to be an efficient tool for exact inference in probabilistic graphical models, by exploiting structures that more traditional methods can not. The availability of affordable high performance commodity hardware has been an inspiration for other inference approaches to exploit parallelism, to great suc- cess. In this paper, we explore the possibilities for Weighted Model Counting. We have empirically confirmed that exploited parallelism yields substantial speedups using a set of real-world Bayesian networks.

AB - Knowledge compilation as part of the Weighted Model Counting approach has proven to be an efficient tool for exact inference in probabilistic graphical models, by exploiting structures that more traditional methods can not. The availability of affordable high performance commodity hardware has been an inspiration for other inference approaches to exploit parallelism, to great suc- cess. In this paper, we explore the possibilities for Weighted Model Counting. We have empirically confirmed that exploited parallelism yields substantial speedups using a set of real-world Bayesian networks.

M3 - Conference contribution

T3 - Proceedings of Machine Learning Research (PMLR)

SP - 97

EP - 108

BT - Proceedings of the Ninth International Conference on Probabilistic Graphical Models

A2 - Kratochvíl, Václav

A2 - Studený, Milan

PB - MLResearchPress

T2 - 9th International Conference on Probabilistic Graphical Models, PGM 2018

Y2 - 11 September 2018 through 14 September 2018

ER -

Parallel probabilistic inference by weighted model counting

Abstract

Publication series

Conference

Access to Document

Fingerprint

Cite this