From Influence Diagrams to Junction Trees
Artificial Intelligence
2013-02-28 v1
Abstract
We present an approach to the solution of decision problems formulated as influence diagrams. This approach involves a special triangulation of the underlying graph, the construction of a junction tree with special properties, and a message passing algorithm operating on the junction tree for computation of expected utilities and optimal decision policies.
Keywords
Cite
@article{arxiv.1302.6824,
title = {From Influence Diagrams to Junction Trees},
author = {Frank Jensen and Finn Verner Jensen and Soren L. Dittmer},
journal= {arXiv preprint arXiv:1302.6824},
year = {2013}
}
Comments
Appears in Proceedings of the Tenth Conference on Uncertainty in Artificial Intelligence (UAI1994)