Updating Probabilities in Multiply-Connected Belief Networks
Jaap Suermondt, Gregory Cooper
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.
PDF Link: /papers/88/p335-suermondt.pdf
AUTHOR = "Jaap Suermondt
and Gregory Cooper",
TITLE = "Updating Probabilities in Multiply-Connected Belief Networks",
BOOKTITLE = "Proceedings of the Fourth Conference Annual Conference on Uncertainty in Artificial Intelligence (UAI-88)",
PUBLISHER = "AUAI Press",
ADDRESS = "Corvallis, Oregon",
YEAR = "1988",
PAGES = "335--343"