[Colloquium] John Langford talk – 10/9 at TTI-C

[Meridel Trimble]

TOYOTA TECHNOLOGICAL INSTITUTE TALK 

Speaker: John Langford 
Toyota Technological Institute at Chicago 
Speaker’s homepage: http://www.tti-c.org/langford.shtml 

Date: Thursday, October 9th, 2003 
Time: 12:15pm 
Place: TTI-C’s Conference Room (The Press Building - 1427 E. 60th St.) 
FREE LUNCH WILL BE PROVIDED 

Title: Robust Approximate Reinforcement Learning

Abstract: In the "real world" the space of an agents' states or observations is 
not practically enumerable, implying that exact algorithms to optimize an 
agents expected reward are impractical.  One common solution is to derive 
approximate forms of the exact algorithms, such as approximate policy 
iteration.  Unfortunately, these approaches tend to be nonrobust since the 
sample complexity (or number of interactions with the world) required to 
guarantee success remains proportional to the size of the state space.Another 
solution (currently being pursued by many people) is to reduce reinforcement 
learning to classification, for which many algorithms and performance 
guarantees are not explicitly or even implicitly dependent on the size of a 
state space.  I will discuss how to do this, what theoretical guarantees can be 
transferred from classification, and show some empirical results suggesting 
this approach works well in practice.

Please contact Meridel Trimble for further information 
(mtrimble at tti-c.org/773.834.9873)