Compiling Relational Bayesian Networks for Exact Inference

Manfred Jaeger; Adnan Darwiche; Mark Chavira

Compiling Relational Bayesian Networks for Exact Inference

Manfred Jaeger, Adnan Darwiche, Mark Chavira

Institut for Datalogi

Publikation: Bidrag til tidsskrift › Tidsskriftartikel › Forskning › peer review

84 Citationer (Scopus)

Abstract

Udgivelsesdato: MAY

Originalsprog	Engelsk
Tidsskrift	International Journal of Approximate Reasoning
Vol/bind	42
Udgave nummer	1-2
Sider (fra-til)	4-20
ISSN	0888-613X
Status	Udgivet - 2006

AUB Link

Søg efter materialet i Aalborg Universitetsbiblioteks søgemaskine

Citationsformater

@article{f42578509fcd11db8ed6000ea68e967b,

title = "Compiling Relational Bayesian Networks for Exact Inference",

abstract = "We describe in this paper a system for exact inference with relational Bayesian networks as defined in the publicly available PRIMULA tool. The system is based on compiling propositional instances of relational Bayesian networks into arithmetic circuits and then performing online inference by evaluating and differentiating these circuits in time linear in their size. We report on experimental results showing successful compilation and efficient inference on relational Bayesian networks, whose PRIMULA--generated propositional instances have thousands of variables, and whose jointrees have clusters with hundreds of variables.",

author = "Manfred Jaeger and Adnan Darwiche and Mark Chavira",

year = "2006",

language = "English",

volume = "42",

pages = "4--20",

journal = "International Journal of Approximate Reasoning",

issn = "0888-613X",

publisher = "Elsevier",

number = "1-2",

}

TY - JOUR

T1 - Compiling Relational Bayesian Networks for Exact Inference

AU - Jaeger, Manfred

AU - Darwiche, Adnan

AU - Chavira, Mark

PY - 2006

Y1 - 2006

N2 - We describe in this paper a system for exact inference with relational Bayesian networks as defined in the publicly available PRIMULA tool. The system is based on compiling propositional instances of relational Bayesian networks into arithmetic circuits and then performing online inference by evaluating and differentiating these circuits in time linear in their size. We report on experimental results showing successful compilation and efficient inference on relational Bayesian networks, whose PRIMULA--generated propositional instances have thousands of variables, and whose jointrees have clusters with hundreds of variables.

AB - We describe in this paper a system for exact inference with relational Bayesian networks as defined in the publicly available PRIMULA tool. The system is based on compiling propositional instances of relational Bayesian networks into arithmetic circuits and then performing online inference by evaluating and differentiating these circuits in time linear in their size. We report on experimental results showing successful compilation and efficient inference on relational Bayesian networks, whose PRIMULA--generated propositional instances have thousands of variables, and whose jointrees have clusters with hundreds of variables.

M3 - Journal article

SN - 0888-613X

VL - 42

SP - 4

EP - 20

JO - International Journal of Approximate Reasoning

JF - International Journal of Approximate Reasoning

IS - 1-2

ER -

Compiling Relational Bayesian Networks for Exact Inference

Abstract

AUB Link

Fingeraftryk

Citationsformater