BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Bayesian Reinforcement Learning - Speaker to be confirmed
DTSTART:20071115T140000Z
DTEND:20071115T153000Z
UID:TALK9210@talks.cam.ac.uk
CONTACT:Sinead Williamson
DESCRIPTION:Milica Gasic and Yunus Saatci will present on Bayesian Reinfor
 cement learning.\n\nReading: A skim of Chapters 3 and 4 of "Sutton and Bar
 to":http://www.cs.ualberta.ca/%7Esutton/book/ebook/the-book.html and a ski
 m of the following papers: "A Bayesian Framework for Reinforcement Learnin
 g (Strens\, 2000)":http://uk.geocities.com/mjstrens/icml.pdf\, "An Analyti
 c Solution to Discrete Bayesian Reinforcement Learning (Poupart et al.\, 2
 006) (up to Section 3.3 only)":http://www.cs.uwaterloo.ca/~ppoupart/public
 ations/bayesian-rl/icml-brl-8pages.pdf\, "Reinforcement Learning with Gaus
 sian Processes (Engel et al.\, 2005)":http://www.machinelearning.org/proce
 edings/icml2005/papers/026_Reinforcement_EngelEtAl.pdf\, "Bayesian Policy 
 Gradient (Ghavamzadeh 2006)":http://www.grappa.univ-lille3.fr/~ppreux/krl/
 Ghavamzadeh-engel.pdf
LOCATION:Engineering Department\, CBL Room 438
END:VEVENT
END:VCALENDAR