An empirical study of Bayesian network inference with simple propagation

Cory J. Butz; Jhonatan Oliveira; Andre E. dos Santos; Anders Læsø Madsen

doi:10.1016/j.ijar.2017.10.005

An empirical study of Bayesian network inference with simple propagation

Cory J. Butz, Jhonatan Oliveira, Andre E. dos Santos, Anders Læsø Madsen

Research output: Contribution to journal › Journal article › Research › peer-review

4 Citations (Scopus)

Abstract

We propose Simple Propagation (SP) as a new join tree propagation algorithm for exact inference in discrete Bayesian networks. We establish the correctness of SP. The striking feature of SP is that its message construction exploits the factorization of potentials at a sending node, but without the overhead of building and examining graphs as done in Lazy Propagation (LP). Experimental results on optimal (or close to optimal) join trees built from numerous benchmark Bayesian networks show that SP is often faster than LP.

Original language	English
Journal	International Journal of Approximate Reasoning
Volume	92
Pages (from-to)	198-211
Number of pages	14
ISSN	0888-613X
DOIs	https://doi.org/10.1016/j.ijar.2017.10.005
Publication status	Published - 1 Jan 2018

Keywords

Bayesian networks
Exact inference
Join tree propagation

Access to Document

10.1016/j.ijar.2017.10.005

AUB Link

Search for the material in Aalborg University Library's search engine

Cite this

@article{458af0234cbc40f699f857093de3bfc8,

title = "An empirical study of Bayesian network inference with simple propagation",

abstract = "We propose Simple Propagation (SP) as a new join tree propagation algorithm for exact inference in discrete Bayesian networks. We establish the correctness of SP. The striking feature of SP is that its message construction exploits the factorization of potentials at a sending node, but without the overhead of building and examining graphs as done in Lazy Propagation (LP). Experimental results on optimal (or close to optimal) join trees built from numerous benchmark Bayesian networks show that SP is often faster than LP.",

keywords = "Bayesian networks, Exact inference, Join tree propagation",

author = "Butz, {Cory J.} and Jhonatan Oliveira and {dos Santos}, {Andre E.} and Madsen, {Anders L{\ae}s{\o}}",

year = "2018",

month = jan,

day = "1",

doi = "10.1016/j.ijar.2017.10.005",

language = "English",

volume = "92",

pages = "198--211",

journal = "International Journal of Approximate Reasoning",

issn = "0888-613X",

publisher = "Elsevier",

}

TY - JOUR

T1 - An empirical study of Bayesian network inference with simple propagation

AU - Butz, Cory J.

AU - Oliveira, Jhonatan

AU - dos Santos, Andre E.

AU - Madsen, Anders Læsø

PY - 2018/1/1

Y1 - 2018/1/1

N2 - We propose Simple Propagation (SP) as a new join tree propagation algorithm for exact inference in discrete Bayesian networks. We establish the correctness of SP. The striking feature of SP is that its message construction exploits the factorization of potentials at a sending node, but without the overhead of building and examining graphs as done in Lazy Propagation (LP). Experimental results on optimal (or close to optimal) join trees built from numerous benchmark Bayesian networks show that SP is often faster than LP.

AB - We propose Simple Propagation (SP) as a new join tree propagation algorithm for exact inference in discrete Bayesian networks. We establish the correctness of SP. The striking feature of SP is that its message construction exploits the factorization of potentials at a sending node, but without the overhead of building and examining graphs as done in Lazy Propagation (LP). Experimental results on optimal (or close to optimal) join trees built from numerous benchmark Bayesian networks show that SP is often faster than LP.

KW - Bayesian networks

KW - Exact inference

KW - Join tree propagation

UR - http://www.scopus.com/inward/record.url?scp=85031923402&partnerID=8YFLogxK

U2 - 10.1016/j.ijar.2017.10.005

DO - 10.1016/j.ijar.2017.10.005

M3 - Journal article

SN - 0888-613X

VL - 92

SP - 198

EP - 211

JO - International Journal of Approximate Reasoning

JF - International Journal of Approximate Reasoning

ER -

An empirical study of Bayesian network inference with simple propagation

Abstract

Keywords

Access to Document

AUB Link

Other files and links

Fingerprint

Cite this