BEGIN:VCALENDAR
VERSION:2.0
CALSCALE:GREGORIAN
PRODID:UW-Madison-Physics-Events
BEGIN:VEVENT
SEQUENCE:1
UID:UW-Physics-Event-6486
DTSTART:20210728T160000Z
DTEND:20210728T171500Z
DTSTAMP:20260414T231231Z
LAST-MODIFIED:20210723T160204Z
LOCATION:Online Seminar: Please sign up for our mailing list at www.ph
 ysicsmeetsml.org for zoom link
SUMMARY:Compositionality of Symmetry in Equivariant Multilayer Percept
 rons\, Physics ∩ ML Seminar\, Siamak Ravanbaksh\, McGill University
DESCRIPTION:A principled approach to modeling structured data is to co
 nsider all transformations that maintain structural relations. Using t
 his perspective in deep learning leads to the design of models that ar
 e invariant or equivariant to the symmetry transformations of the data
 . Symmetry groups can be composed in different ways\, and in this talk
 \, I explore the utility of the compositionality of symmetry groups in
  deep learning. In particular\, I review different notions of group co
 mposition and their application in deep model design for new and compo
 sitional data structures.
URL:https://www.physics.wisc.edu/events/?id=6486
END:VEVENT
END:VCALENDAR