BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4//
BEGIN:VEVENT
UID:20260624T020729EDT-6794RvlcAP@132.216.98.100
DTSTAMP:20260624T060729Z
DESCRIPTION:\n	Abstract:\n\n\nSymmetry has played a significant role in mode
 rn physics\, in part by constraining the physical laws. I will discuss how
  it could play a fundamental role in AI by constraining the deep model des
 ign. In particular\, I focus on discrete domain symmetries and through exa
 mples show how we can use this inductive bias as a principled means for co
 nstraining a feedforward layer and significantly improving its sample effi
 ciency.\n\n\n	Speaker\n\n\nSiamak Ravanbakhsh is an Assistant Professor\, S
 chool of Computer Science\, 91ºÚÁÏÍø. His research interests incl
 ude inference within structured\, complex and combinatorial domains using 
 graphical and structured deep models.\n\nSeminar website:\n\n \n\nhttps://
 mcgillstat.github.io\n
DTSTART:20190920T193000Z
DTEND:20190920T203000Z
LOCATION:Room 1205\, Burnside Hall\, CA\, QC\, Montreal\, H3A 0B9\, 805 rue
  Sherbrooke Ouest
SUMMARY:Siamak Ravanbakhsh (91ºÚÁÏÍø - Computer Science)
URL:/mathstat/channels/event/siamak-ravanbakhsh-mcgill
 -computer-science-300721
END:VEVENT
END:VCALENDAR