BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:IC Colloquium: Progressive growing of GANs
DTSTART:20181217T161500
DTEND:20181217T173000
DTSTAMP:20260510T091530Z
UID:2f2f77b4d1a69f9b75deee5676c17a4fbd097c66ca3bafc0b95b12bc
CATEGORIES:Conferences - Seminars
DESCRIPTION:By: Jaakko Lehtinen - Aalto University\n\nAbstract:\nGenerativ
 e adversarial networks (GAN) are a powerful and exciting family of generat
 ive models that learn purely from observing samples. The quality of sample
 s has\, however\, remained less than ideal. We describe a new GAN training
  methodology that yields samples of unprecedented quality. The key idea is
  to grow both the generator and discriminator progressively: starting from
  a low resolution\, we add new layers that model increasingly fine details
  as training progresses. A form of curriculum learning\, this both speeds 
 the training up and greatly stabilizes it. We also propose a simple way to
  increase the variation in generated images. This talk will also feature s
 till higher-quality results than presented in the ICLR article.\n\nBio:\nJ
 aakko Lehtinen is an associate professor at Aalto University\, Finland\, a
 nd a principal research scientist with NVIDIA Research. Prior to that\, h
 e spent a few years as a postdoc with Frédo Durand at MIT. Jaakko works
  in the intersection of computer graphics and computer vision\, including
  imaging and generative modelling. Of the latter line of research\, Progre
 ssively-grown GANs* have recently gathered significant attention.\n\nMore 
 information
LOCATION:BC 420 https://plan.epfl.ch/?room==BC%20420
STATUS:CANCELLED
END:VEVENT
END:VCALENDAR