BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:The Bayesian Approach to Inverse Problems: Computational Aspects -
  Felix Lucka (University of Münster)
DTSTART:20121114T150000Z
DTEND:20121114T160000Z
UID:TALK40781@talks.cam.ac.uk
CONTACT:Carola-Bibiane Schoenlieb
DESCRIPTION:In the second talk\, I want to discuss some practical aspects 
 of Bayesian inference applied to inverse problems. As an exemplary applica
 tion\, I will consider solving high-dimensional inverse problems using spa
 rsity constraints as a-priori information\, e.g.\, the well-known total va
 riation (TV) minimization constraint. After explaining the basic principle
 s and algorithms of Markov chain Monte Carlo (MCMC) based posterior infere
 nce\, I will show that contrary to what is commonly believed about the app
 licability of MCMC schemes\, they are not in general slow and scale bad wi
 th increasing dimension. In addition\, I will outline why I think that de
 signing efficient optimization and sampling techniques is conceptually sim
 ilar\, and highlight some recent work on enhancing sampling by incorporati
 ng ideas from optimization.\n\nSlides of this talk can be downloaded at ht
 tp://wwwmath.uni-muenster.de/num/burger/organization/lucka/talks/Cambridge
 2_14_11_2012.pdf
LOCATION:CMS\, MR5
END:VEVENT
END:VCALENDAR