BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Epitomes And Epitexts - Phil Cowans\, Inference Group
DTSTART:20050614T130000Z
DTEND:20050614T140000Z
UID:TALK4327@talks.cam.ac.uk
CONTACT:Phil Cowans
DESCRIPTION:The epitome model of images\, originally proposed by Jojic and
  Frey divides the image into overlapping patches\, each of which is genera
 ted probabilistically from a smaller image known as the epitome. The epito
 me therefore represents local texture in the original image. Applications 
 of this model include reconstrustion of occluded pixels in video and image
  super-resolution. \n\nIn this talk I will describe our application of the
  same technique to text modelling. For this task\, fragments of the text b
 eing modelled are drawn from distributions defined by the equivalent of th
 e epitome\, the epitext. This can be viewed as a generalisation of the 'to
 pics' used in models such as latent Dirichlet allocation.\n\nThis is joint
  work with John Winn.
LOCATION:Ryle Seminar Room\, Cavendish Laboratory
END:VEVENT
END:VCALENDAR