BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Deep Generative Models - Jonathan Gordon\; Alexander Matthews
DTSTART:20180201T133000Z
DTEND:20180201T150000Z
UID:TALK94237@talks.cam.ac.uk
CONTACT:Shixiang Gu
DESCRIPTION:Deep\, latent variable generative models have recently attract
 ed a lot of attention from the research community. In the first half of th
 is talk we will present a number of recent advances in variational auto-en
 coder (VAE) based models\, focusing on the use of probabilistic modelling 
 tools to increase the\ncapacity of the standard VAE and allowing it to be 
 applied to a broader range of learning tasks. The talk will be self contai
 ned\, but will only include a brief review of the standard VAE concepts. I
 t is recommended therefore to be familiar with the concepts presented in [
 1].\n\n[1] - Kingma D P\, and Welling M. Auto-Encoding Variational Bayes [
 1]\n\nThe second half of the talk will cover a subset of the literature on
  tractable deep generative models\, which can be trained by regularized ma
 ximum likelihood. We will cover auto-regressive models such as Pixel Convo
 lutional Networks and models composed of a sequence of tractable transform
 ations such as Real NVP. We will compare and contrast the various approach
 es discussed.\n
LOCATION:Engineering Department\, CBL Seminar Room 4-38
END:VEVENT
END:VCALENDAR