BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Bayesian analysis and computation for convex inverse problems: the
 ory\, methods\, and algorithms - Marcelo Pereyra (Heriot-Watt University)
DTSTART:20171102T145000Z
DTEND:20171102T154000Z
UID:TALK94363@talks.cam.ac.uk
CONTACT:INI IT
DESCRIPTION:This talk presents some new developments in theory\, methods\,
  and algorithms for performing Bayesian inference in high-dimensional inve
 rse problems that are convex\, with application to mathematical and comput
 ational imaging. These include new efficient stochastic simulation and opt
 imisation Bayesian computation methods that tightly combine proximal optim
 isation with Markov chain Monte Carlo techniques\; strategies for estimati
 ng unknown model parameters and performing model selection\, methods for c
 alculating Bayesian confidence intervals for images and performing uncerta
 inty quantification analyses\; and new theory regarding the role of convex
 ity in maximum-a-posteriori and minimum-mean-square-error estimation. The 
 new theory\, methods\, and algorithms are illustrated with a range of math
 ematical imaging experiments.
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR