BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:How to force unsupervised neural networks to discover the right re
 presentation of images - Geoffrey Hinton\, University of Toronto
DTSTART:20110623T100000Z
DTEND:20110623T110000Z
UID:TALK31612@talks.cam.ac.uk
CONTACT:Microsoft Research Cambridge Talks Admins
DESCRIPTION:One appealing way to design an object recognition system is to
  define objects recursively in terms of their parts and the required spati
 al relationships between the parts and the whole. These relationships can 
 be represented by the coordinate transformation between an intrinsic frame
  of reference embedded in the part and an intrinsic frame embedded in the 
 whole. This transformation is unaffected by the viewpoint so this form of 
 knowledge about the shape of an object is viewpoint invariant. A natural w
 ay for a neural network to implement this knowledge is by using a matrix o
 f weights to represent each part-whole relationship and a vector of neural
  activities to represent the pose of each part or whole relative to the vi
 ewer. The pose of the whole can then be predicted from the poses of the pa
 rts and\, if the predictions agree\, the whole is present. This leads to n
 eural networks that can recognize objects over a wide range of viewpoints 
 using neural activities that are ``equivariant'' rather than invariant: as
  the viewpoint varies the neural activities all vary even though the knowl
 edge is viewpoint-invariant. The ``capsules'' that implement the lowest-le
 vel parts in the shape hierarchy need to extract explicit pose parameters 
 from pixel intensities and these pose parameters need to have the right fo
 rm to allow coordinate transformations to be implemented by matrix multipl
 ies. These capsules are quite easy to learn from pairs of transformed imag
 es if the neural net has direct\, non-visual access to the transformations
 \, as it would if it controlled them. (Joint work with Sida Wang and Alex 
 Krizhevsky)
LOCATION:Small lecture theatre\, Microsoft Research Ltd\, 7 J J Thomson Av
 enue (Off Madingley Road)\, Cambridge
END:VEVENT
END:VCALENDAR