BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Learning Logical Relations - Christian Steinruecken (University of
  Cambridge)
DTSTART:20091012T100000Z
DTEND:20091012T110000Z
UID:TALK19957@talks.cam.ac.uk
CONTACT:Emli-Mari Nel
DESCRIPTION:Given some (possibly incomplete) knowledge about features of g
 iven a set of objects\, it is possible to construct probabilistic models w
 hich compress the data by learning an underlying structural organisation (
 such as a hierarchy).\nIn addition to structural form\, however\, real-wor
 ld datasets often exhibit logical constraints which\, when discovered\, ca
 n be exploited to better compress and predict the data.\nI will discuss th
 e construction of models which capture such latent logical relations\, and
  how this can be integrated into existing structure-discovering models.\n\
 nThis is early work and many things aren't quite settled yet.  Questions a
 re welcome and I'd be grateful for ideas and feedback.
LOCATION:TCM Seminar Room\, Cavendish Laboratory\, Department of Physics
END:VEVENT
END:VCALENDAR