BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Machine learning based approaches for decision & control under unc
 ertainties (in electric power systems) - Professor Louis Wehenkel (Departm
 ent of Electrical Engineering and Computer Science\, University of Liege)
DTSTART:20081112T110000Z
DTEND:20081112T120000Z
UID:TALK15103@talks.cam.ac.uk
CONTACT:Dr Guy-Bart Stan
DESCRIPTION:The talk will present two approaches for designing sequential 
 decision strategies by exploiting ideas from supervised learning. The firs
 t one is a non-model based batch-mode reinforcement learning algorithm ite
 ratively fitting Q-functions to a random sample of four-tuples (state\, ac
 tion\, reward\, next-state). The second one is a model based multi-stage s
 tochastic programming algorithm extracting and aggregating optimized decis
 ions computed over an ensemble of incomplete disturbance trees.  The algor
 ithms are motivated by and illustrated on electric power systems planning 
 and control problems.
LOCATION:Cambridge University Engineering Department\, Lecture Room 6
END:VEVENT
END:VCALENDAR