Efficiently Learning to Behave Efficiently

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
within a novel machine learning framework we call "kWIK" for "knows
what it knows" . I'll summarize the framework, our algorithms, their
formal validations, and their empirical evaluations in robotic and
videogame testbeds.
Michael L. Littman directs the Rutgers Laboratory for Real Life
Reinforcement Learning (RL3) and his research in machine learning
examines algorithms for decision making under uncertainty. Littman
worked as an Assistant Professor at Duke University, a Member of
Technical Staff in AT&T's Artificial Intelligence Principles Research
Department, an Adjunct Professor at Princeton University, and is now a
Professor of Computer Science at Rutgers University. Both Duke and
Rutgers honored him with undergraduate teaching awards and his
research has been recognized with five best-paper awards on topics
ranging from algorithms for efficient reinforcement learning and
sequential decision making, to computational game theory, to computer
crossword solving. He has served as associate editor for three of the
major journals in his field.