BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Advanced MCMC Methods - Iain Murray\, Gatsby Unit\, UCL
DTSTART:20061123T160000Z
DTEND:20061123T180000Z
UID:TALK5543@talks.cam.ac.uk
CONTACT:Zoubin Ghahramani
DESCRIPTION:Markov chain Monte Carlo (MCMC) algorithms draw correlated sam
 ples from probability distributions. These allow approximate computation o
 f complex high-dimensional integrals\; obvious applications include Bayesi
 an statistics and statistical physics. This tutorial will not assume any p
 rior knowledge of MCMC\, but will cover state-of-the-art techniques.\n\nTe
 ntative Schedule:\n\n* Introduction: Metropolis--Hastings vs "simpler" Mon
 te Carlo methods\n* Hamiltonian (Hybrid) Monte Carlo\n* Auxiliary variable
 s and Slice sampling\n* Out of equilibrium: tempering/annealing and relate
 d advances.\n* Possibly a few words on infinite models and doubly-intracta
 ble distributions.
LOCATION:LR4\, Engineering\, Department of
END:VEVENT
END:VCALENDAR