BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Autoregressive Diffusion Models - Emiel Hoogeboom
DTSTART:20220330T100000Z
DTEND:20220330T113000Z
UID:TALK172085@talks.cam.ac.uk
CONTACT:Elre Oldewage
DESCRIPTION:Autoregressive diffusion models are a new generative model tha
 t can be seen as an autoregressive model that places a distribution over v
 ariables in random dimension order.   In this talk we will discuss their r
 elation to standard (Gaussian-based) diffusion models\, and the lessons (p
 ositive and negative) we can draw from this project.\n\nRecommended readin
 g:\nthe original ‘modern’ diffusion paper (DDPMs) at https://arxiv.org
 /abs/2006.11239\n\nBio:\nEmiel is a last year PhD candidate at the Univers
 ity of Amsterdam under the supervision of Max Welling. Research interests 
 include generative modelling and variational inference\, for instance focu
 sing on discrete variable extensions. He completed a BSc in Aerospace Engi
 neering and a MSc in Artificial Intelligence. After a 3 month sabbatical h
 e will join Google Brain in Amsterdam.
LOCATION:Cambridge University Engineering Department\, CBL Seminar room BE
 4-38
END:VEVENT
END:VCALENDAR