Learning Link-Probabilities in Causal Trees
Igor Roizer, Judea Pearl
A learning algorithm is presented which given the structure of a causal tree, will estimate its link probabilities by sequential measurements on the leaves only. Internal nodes of the tree represent conceptual (hidden) variables inaccessible to observation. The method described is incremental, local, efficient, and remains robust to measurement imprecisions.
PDF Link: /papers/86/p211-roizer.pdf
AUTHOR = "Igor Roizer
and Judea Pearl",
TITLE = "Learning Link-Probabilities in Causal Trees",
BOOKTITLE = "Proceedings of the Second Conference Annual Conference on Uncertainty in Artificial Intelligence (UAI-86)",
PUBLISHER = "AUAI Press",
ADDRESS = "Corvallis, Oregon",
YEAR = "1986",
PAGES = "211--214"