BEGIN:VCALENDAR
VERSION:2.0
PRODID:researchseminars.org
CALSCALE:GREGORIAN
X-WR-CALNAME:researchseminars.org
BEGIN:VEVENT
SUMMARY:Iordan Ganev (Weizmann)
DTSTART:20210315T180000Z
DTEND:20210315T190000Z
DTSTAMP:20260423T004642Z
UID:UMassRep/19
DESCRIPTION:Title: <a href="https://researchseminars.org/talk/UMassRep/19/
 ">The QR decomposition for radial neural networks.</a>\nby Iordan Ganev (W
 eizmann) as part of UMass Amherst Representation theory seminar\n\n\nAbstr
 act\nWe present a perspective on neural networks stemming from quiver repr
 esentation theory. This point of view emphasizes the symmetries inherent i
 n neural networks\, interacts nicely with gradient descent\, and has the p
 otential to improve training algorithms. As an application\, we establish 
 an analog of the QR decomposition for radial neural networks\, which leads
  to a dimensional reduction result. This talk is intended for an audience 
 with a background in representation theory\; we explain all concepts relat
 ing to neural networks and machine learning from first principles. It is b
 ased on joint work with Robin Walters.\n
LOCATION:https://researchseminars.org/talk/UMassRep/19/
END:VEVENT
END:VCALENDAR