Title: Local Distributions in Bayesian Networks: Knowledge Engineering and Inference

When: Monday, September 27, 2004 12:00 noon

Where: Room 501, IS Building

Who: Adam Zagorecki, PhD Student, Information Science Program

Abstract: Bayesian networks are a popular tool for modeling uncertainty that has a solid record of successful real world applications. One of the major limitations of Bayesian networks which is preventing from building large models, is an exponential growth of parameters required to specify local probability distributions. A number of various solutions have been proposed to address this problem. In this talk, I present a review of the current literature addressing the problem of compact representations of local probability distributions in the Bayesian network framework. This talk is intended to be a dry-run for my comprehensive examination.

Speaker Bio
*************
Adam Zagorecki is a Ph.D. student in the Department of Information Science and Telecommunications, University of Pittsburgh. He is a member of the Decision Systems Laboratory, which specializes in applying decision-theoretic techniques in decision support. His research interests include causal independence in Bayesian networks, modeling and diagnosis.

Email : adamz@sis.pitt.edu
Website: www.sis.pit.edu/~adamz