BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Variational autoencoders with latent graphical models - Prof David
  Duvenaud (University of Toronto)
DTSTART:20161213T110000Z
DTEND:20161213T120000Z
UID:TALK67586@talks.cam.ac.uk
CONTACT:Zoubin Ghahramani
DESCRIPTION:We propose a general modeling and inference framework that com
 poses probabilistic graphical models with deep learning methods\, in a way
  that combines their respective strengths. Our model family combines graph
 ical structure in latent variables with neural network observation models.
  For inference\, we use variational recognition networks to produce  local
  evidence summaries\, and combine them using exact graphical model inferen
 ce.  We illustrate this framework with several example models\, and an app
 lication to automatic mouse behavior modeling from video.
LOCATION:CBL Room BE-438\, Department of Engineering
END:VEVENT
END:VCALENDAR