BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Collective sampling through a Metropolis-Hasting like method: kine
 tic theory and numerical experiments - Antoine Diez\, Imperial College Lon
 don and Grégoire Clarté\, Université Paris Dauphine 
DTSTART:20191204T160000Z
DTEND:20191204T170000Z
UID:TALK135301@talks.cam.ac.uk
CONTACT:84031
DESCRIPTION:The classical Metropolis-Hastings algorithm provides a simple 
 method to construct a Markov Chain with an arbitrary stationary measure. I
 n order to implement Monte Carlo methods\, an elementary approach would be
  to duplicate this algorithm as many times as desired. Following the ideas
  of Population Monte Carlo methods\, we propose to take advantage of the n
 umber of duplicates to increase the efficiency of the naive approach. With
 in this framework\, each chain is seen as the evolution of a single partic
 le which interacts with the others. In this article\, we propose a simple 
 and efficient interaction mechanism and an analytical framework which ensu
 res that the particles are asymptotically independent and identically dist
 ributed according to an arbitrary target law. This approach is also suppor
 ted by numerical simulations showing better convergence properties compare
 d to the classical Metropolis-Hastings algorithm.\n
LOCATION:MR14\, Centre for Mathematical Sciences
END:VEVENT
END:VCALENDAR