BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:A Bayesian approach to multicanonical Markov chain Monte Carlo - D
 r Jes Frellsen\, CBL\, CUED
DTSTART:20150129T140000Z
DTEND:20150129T150000Z
UID:TALK57241@talks.cam.ac.uk
CONTACT:Fredrik Lindsten
DESCRIPTION:The Markov chain Monte Carlo method is one of the most importa
 nt tools for approximate inference. However\, the canonical ensemble\, kno
 wn as the posterior distribution in Bayesian inference and the Boltzmann d
 istribution in physics\, is in many cases difficult to sample from due to 
 slow convergence and poor mixing of the Markov chain. The so-called multic
 anonical ensemble can potentially alleviate this problem by constructing a
  new distribution that performs a random walk between different values of 
 the energy function. Estimates for posterior averages and partition functi
 ons can then be obtained by reweighting. The main challenge in applying th
 e multicanonical ensemble is that one has to solve an inference problem on
  the density of energy under the prior. In this talk I will present a Baye
 sian approach to solve this problem using a Gaussian Process prior. I will
  present preliminary results on spin models and discuss applications in pr
 otein folding.
LOCATION:Board Room\, CUED
END:VEVENT
END:VCALENDAR