BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Entropy contraction of the Gibbs sampler under log-concavity - Gia
 como Zanella (Bocconi University)
DTSTART:20241115T140000Z
DTEND:20241115T150000Z
UID:TALK222256@talks.cam.ac.uk
CONTACT:Qingyuan Zhao
DESCRIPTION:In this talk I will present recent work (https://arxiv.org/abs
 /2410.00858) on the non-asymptotic analysis of the Gibbs sampler\, which i
 s a canonical and popular Markov chain Monte Carlo algorithm for sampling.
  In particular\, under the assumption that the probability measure π of i
 nterest is strongly log-concave\, we show that the random scan Gibbs sampl
 er contracts in relative entropy and provide a sharp characterization of t
 he associated contraction rate. The result implies that\, under appropriat
 e conditions\, the number of full evaluations of π required for the Gibbs
  sampler to converge is independent of the dimension. If time permits\, I 
 will also discuss connections and applications of the above results to the
  problem of zero-order parallel sampling. \n\nBased on joint work with Fil
 ippo Ascolani and Hugo Lavenant.
LOCATION:Centre for Mathematical Sciences MR12\, CMS
END:VEVENT
END:VCALENDAR