Distinguished Lecture Series

 

"Efficiently Learning to Behave Efficiently"

Michael Littman, Rutgers University

Thursday, October 29, 2009 at 4:00 P.M.

Room 368 (CIT 3rd floor)

The field of reinforcement learning is concerned with the problem of learning efficient behavior from experience. In real life applications, gathering this experience is time-consuming and possibly costly, so it is critical to derive algorithms that can learn effective behavior with bounds on the experience necessary to do so. This talk presents our successful efforts to create such algorithms via a new framework we call KWIK (Knows What It Knows) learning. I'll summarize the framework, our algorithms, their formal validations, and their empirical evaluations in robotic and videogame testbeds. This approach holds promise for attacking challenging problems in a number of application domains.

A reception will follow in the atrium.

Host: John Hughes

To watch part one of this lecture, please click here.

To watch part two of this lecture, please click here.

To watch part three of this lecture, please click here.

To watch part four of this lecture, please click here.

To watch part five of this lecture, please click here.

To watch part six of this lecture, please click here.

To watch part seven of this lecture, please click here.