BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Nested Sampling: an efficient and robust Bayesian inference tool f
 or Machine Learning and Data Science - Will Handley - Kavli
DTSTART:20191114T130000Z
DTEND:20191114T143000Z
UID:TALK131608@talks.cam.ac.uk
CONTACT:James Fergusson
DESCRIPTION:Nested sampling is an MCMC technique for integrating and explo
 ring probability distributions. It has become widely adopted in the field 
 of cosmology as a powerful tool for computing Bayesian evidences and sampl
 ing challenging a-priori unknown parameter spaces.\n\nIn this talk\, I wil
 l give an introduction to the principles of Bayesian model comparison and 
 parameter estimation\, an explanation of the theory of nested sampling\, a
  survey of the current state-of-the art (MultiNest\, PolyChord\, DNest and
  Dynesty) and the future of the field. I will illustrate with applications
  in CMB and 21cm Cosmology\, Bayesian Sparse Reconstruction and Bayesian N
 eural Networks.
LOCATION:Kavli Large Meeting Room\, Kavli Building
END:VEVENT
END:VCALENDAR