Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!uunet!samsung!usc!ucla-cs!maui.cs.ucla.edu!gblee From: gblee@maui.cs.ucla.edu (Geunbae Lee) Newsgroups: comp.ai Subject: Re: info wanted on learning probabilities Keywords: learning, probability, conditional probabilities Message-ID: <1991Jan30.213055.25485@cs.ucla.edu> Date: 30 Jan 91 21:30:55 GMT References: Sender: news@cs.ucla.edu (Shemp News Account) Organization: UCLA Computer Science Department Lines: 21 Nntp-Posting-Host: maui.cs.ucla.edu I think the following book and other Pearl's works can answer your question (sorry for latex format). Pearl's Bayesian network can formulate, propagate, and revise beliefs (similar to your conditional probablility) according to prior experiences. The book has a good bibliography for this field too. @Book{pearl:probabilistic, author = "Judea Pearl", title = "Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference", publisher = KAUF, year = "1988", address = KAUF-ADDR, rem = "pearl.88b" } -- +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ + Geunbae Lee, Artificial Intelligence Lab, Computer Science Dept, UCLA. + + INTERNET:gblee@cs.ucla.edu, PHONE:213-825-5199 (office) + + Sir, AI is the science that makes machines smart, but people dumb!!! +