• Event Date: October 22, 2007
  • Event Start Time: 1:00 PM
  • Event End Time: 2:00 PM
  • Event Location: Rutgers University, Department of Computer Science
  • Event Type: Human and Computer Vision Series
  • Event Semester: Fall 2007
  • Event Contact: Dr. Michael Littman
  • Event Extra info: <a href="http://www.cs.rutgers.edu/~mlittman/">Dr. Michael Littman</a>

This talk addresses the problem of learning efficiently to make
sequential decisions with a particular focus on generalizing
experience without forfeiting formal learning-time guarantees.  I'll
summarize the theoretically motivated algorithms my group has been
developing that exhibit practical advantages over existing learning
algorithms.  I'll also toss in some video footage of robots learning
to move around by exploring efficiently.

Background Reading:

Reinforcement Learning:  A Survey