Updating Probabilities in Multiply-Connected Belief Networks
Abstract
This paper focuses on probability updates in multiply-connected belief networks. Pearl has designed the method of conditioning, which enables us to apply his algorithm for belief updates in singly-connected networks to multiply-connected belief networks by selecting a loop-cutset for the network and instantiating these loop-cutset nodes. We discuss conditions that need to be satisfied by the selected nodes. We present a heuristic algorithm for finding a loop-cutset that satisfies these conditions.
Cite
Text
Suermondt and Cooper. "Updating Probabilities in Multiply-Connected Belief Networks." Conference on Uncertainty in Artificial Intelligence, 1988.Markdown
[Suermondt and Cooper. "Updating Probabilities in Multiply-Connected Belief Networks." Conference on Uncertainty in Artificial Intelligence, 1988.](https://mlanthology.org/uai/1988/suermondt1988uai-updating/)BibTeX
@inproceedings{suermondt1988uai-updating,
title = {{Updating Probabilities in Multiply-Connected Belief Networks}},
author = {Suermondt, Jaap and Cooper, Gregory F.},
booktitle = {Conference on Uncertainty in Artificial Intelligence},
year = {1988},
url = {https://mlanthology.org/uai/1988/suermondt1988uai-updating/}
}