BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Machine Learning on Sets -  (University of Cambridge)
DTSTART:20200408T100000Z
DTEND:20200408T113000Z
UID:TALK141466@talks.cam.ac.uk
CONTACT:75379
DESCRIPTION:For modality such as image and sequence(text\, audio)\, the in
 put/output ordering contains information which will lost after performing 
 random permutation on the data. However\, in several other domains such as
  (sub) graphs and 3D meshes/point clouds\, it is more natural to represent
  each instance as the set of its components or parts. Many conventional ma
 chine learning algorithms are unable to process this kind of representatio
 ns\, since sets may vary in cardinality and elements lack a meaningful ord
 ering. In this talk\, we will present a paper introducing a framework call
 ed ‘Deep Sets’ to deal with such set representations\, whose permutati
 on invariance has been proved necessarily and sufficiently. Furthermore\, 
 we will discuss some specific permutation invariant/equivariant (model & l
 oss) designs on 3D Vision and Graphs.
LOCATION:https://meet.google.com/qvg-knkr-nrg
END:VEVENT
END:VCALENDAR