BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Deep Gaussian Process Priors for Bayesian Inverse Problems - Areth
 a  Teckentrup (University of Edinburgh)
DTSTART:20180412T103000Z
DTEND:20180412T110000Z
UID:TALK103720@talks.cam.ac.uk
CONTACT:INI IT
DESCRIPTION:Co-authors: Matt Dunlop (Caltech)\, Mark Girolami (Imperial Co
 llege)\, Andrew Stuart (Caltech)<br><br>Deep Gaussian processes have recei
 ved a great deal of attention in the last couple of years\, due to their a
 bility to model very complex behaviour. In this talk\, we present a genera
 l framework for constructing deep Gaussian processes\, and provide a mathe
 matical argument for why the depth of the processes is in most cases finit
 e. We also present some numerical experiments\, where deep Gaussian proces
 ses have been employed as prior distributions in Bayesian inverse problems
 .<br><br>Related Links<ul><li><a target="_blank" rel="nofollow">https://ar
 xiv.org/abs/1711.11280</a>&nbsp\;- Preprint</li></ul><br>
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR