On the structure of elimination trees for Bayesian network inference

Authors:
Kevin Grant;Keilan Scholten
Affiliations:
Department of Math and Computer Science, University of Lethbridge, Lethbridge, Alberta, Canada;Department of Math and Computer Science, University of Lethbridge, Lethbridge, Alberta, Canada
Venue:
MICAI'10 Proceedings of the 9th Mexican international conference on Artificial intelligence conference on Advances in soft computing: Part II
Year:
2010

Citing 7
Cited 0

The computational complexity of probabilistic inference using Bayesian belief networks (research note)

Artificial Intelligence
Computational intelligence: a logical approach

Computational intelligence: a logical approach
Recursive conditioning

Artificial Intelligence - special issue on computational tradeoffs under bounded resources
Using Recursive Decomposition to Construct Elimination Orders, Jointrees, and Dtrees

ECSQARU '01 Proceedings of the 6th European Conference on Symbolic and Quantitative Approaches to Reasoning with Uncertainty
Methods for constructing balanced elimination trees and other recursive decompositions

International Journal of Approximate Reasoning
New advances in inference by recursive conditioning

UAI'03 Proceedings of the Nineteenth conference on Uncertainty in Artificial Intelligence
Conditioning graphs: practical structures for inference in bayesian networks

AI'05 Proceedings of the 18th Australian Joint conference on Advances in Artificial Intelligence

Quantified Score

Hi-index	0.00

Visualization

Abstract

We present an optimization to elimination tree inference in Bayesian networks through the use of unlabeled nodes, or nodes that are not labeled with a variable from the Bayesian network. Through the use of these unlabeled nodes, we are able to restructure these trees, and reduce the amount of computation performed during the inference process. Empirical tests show that the algorithm can reduce multiplications by up to 70%, and overall runtime by up to 50%.