Broad Area Colloquium For AI-Geometry-Graphics-Robotics-Vision


Learning Bayesian Networks

David Heckerman
Senior Researcher
Machine Learning and Applied Statistics Group
Microsoft Research

Wednesday, January 19, 2000
refreshments 4:05PM, talk begins 4:15PM
TCseq201, Lecture Hall B
http://robotics.stanford.edu/ba-colloquium/

Abstract

For two decades, Bayesian networks--also known as directed acyclic graphical models, belief networks, and causal networks--have been used in intelligent systems with a fair amount of success. With few exceptions, system builders have constructed Bayesian networks by directly encoding the knowledge of experts. Data sets have rarely been used in the construction process. One drawback of this knowledge-based approach is that knowledge elicitation can be expensive. More recently, however, researchers have developed techniques for constructing Bayesian networks (both parameters and structure) from a combination of expert knowledge and data. These methods can significantly reduce the cost of building an intelligent system in domains where data is readily available. In addition, these techniques can be used to identify causal relationships from non-experimental data--an important breakthrough for science. I my talk, I will describe several applications of this work being addressed at Microsoft and briefly review the technology.

About the Speaker

David Heckerman is a Senior Researcher at Microsoft Research and manager of the Machine Learning and Applied Statistics Group. He is a co-creator of Answer Wizard for Office 95, Office Assistant for Office 97, troubleshooters for online support and Windows 98, and Intelligent Cross Sell for Commerce Server. Heckerman received his Ph.D. and M.D. degrees from Stanford University.
bac-coordinators@cs.stanford.edu
Back to the Colloquium Page
Last modified: Mon Jan 17 19:33:36 PST 2000