BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Optimal Control and Reinforcement Learning with Gaussian Process M
 odels - Marc Deisenroth (University of Cambridge)
DTSTART:20071107T113000Z
DTEND:20071107T123000Z
UID:TALK9150@talks.cam.ac.uk
CONTACT:Zoubin Ghahramani
DESCRIPTION:Optimal control and reinforcement learning (RL) have the same 
 objective: optimization of a long-term performance measure. While the syst
 em in optimal control problems is usually known\, RL has a more general se
 tup\, which includes possibly unknown environments. However\, after learni
 ng a model standard algorithms for optimal control can also be applied to 
 RL.\n\nIn this talk a generalization of dynamic programming (DP) to contin
 uous-valued state and action spaces is given. The proposed algorithm (GPDP
 ) combines Gaussian process (GP) models with DP and yields an approximate 
 optimal closed-loop policy on the entire state space. We apply GPDP to the
  underactuated pendulum swing up. For exactly known environments we show t
 hat GPDP yields an close-to optimal solution. Moreover\, we show that GPDP
  can successfully be applied to stochastic optimal control problems.
LOCATION:Engineering Department\, CBL Room 438
END:VEVENT
END:VCALENDAR