The Bayesian Network Story
Richard E. Neapolitan, Xia Jiang · Oxford University Press eBooks · 2017
Bayesian networks are now among the leading architectures for reasoning with uncertainty in artificial intelligence. This chapter concerns their story, namely what they are, how and why they came into being, how we obtain them, and what they actually represent. First, it is shown that a standard application of Bayes’ Theorem constitutes inference in a two-node Bayesian network. Then more complex Bayesian networks are presented. Next the genesis of Bayesian networks and their relationship to causality is presented. A technique for learning Bayesian networks from data follows. Finally, a discussion of the philosophy of the probability distribution represented by a Bayesian network is provided.