BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Adaptive Hamiltonian-based MCMC samplers - Shakir Mohamed 
DTSTART:20130523T140000Z
DTEND:20130523T153000Z
UID:TALK45332@talks.cam.ac.uk
CONTACT:Colorado Reed
DESCRIPTION:In this RCC we discuss the widely-experienced difficulty in tu
 ning Monte Carlo samplers based on simulating Hamiltonian dynamics. We dev
 elop an algorithm that allows for the adaptation of Hamiltonian and Rieman
 n manifold Hamiltonian Monte Carlo samplers using Bayesian optimization th
 at allows for infinite adaptation of the parameters of these samplers. We 
 show that the resulting samplers are ergodic\, and that the use of our ada
 ptive algorithms makes it easy to obtain more efficient samplers\, in some
  cases precluding the need for more complex solutions. Hamiltonian-based M
 onte Carlo samplers are widely known to be an excellent choice of MCMC met
 hod\, and such approaches remove a key obstacle towards the more widesprea
 d use of these samplers in practice.
LOCATION:Engineering Department\, CBL Room 438
END:VEVENT
END:VCALENDAR