BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Implicit Variational Inference - Andrew Foong (University of Cambr
 idge)
DTSTART:20190313T134500Z
DTEND:20190313T151500Z
UID:TALK121618@talks.cam.ac.uk
CONTACT:75379
DESCRIPTION:\nVariational inference is frequently the preferred method for
  modelling complicated posteriors arising from Bayesian inference. Classic
 al variational methods restrict the approximate posterior to the exponenti
 al family\, which may lead to large  amounts of bias in the estimation of 
 model parameters. Many methods have been devised in recent years for allow
 ing more flexible posteriors. In this talk\, we will discuss recent advanc
 es in variational inference with implicit distributions: distributions fro
 m which we can draw samples and compute gradients\, but do not have analyt
 ic expression for.\n\nSuggested reading:\nAdversarial Variational Bayes - 
 Mescheder et. al. 2017\nKernel Implicit Variational Inference - Shi et. al
 . 2018\nSemi-Implicit Variational Inference -  Yin & Zhou 2018\nUnbiased I
 mplicit Variational Inference - Titsias & Ruiz 2019
LOCATION:Engineering Department\, CBL Room BE-438
END:VEVENT
END:VCALENDAR