BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4//
BEGIN:VEVENT
UID:20260917T114522EDT-3516McpZPH@132.216.98.100
DTSTAMP:20260917T154522Z
DESCRIPTION:Title: Gradient flows in the deep linear network.\n\nAbstract: 
 The deep linear network (DLN) is a phenomenological random matrix model of
  deep learning that was introduced by Arora\, Cohen and Hazan in 2018. Thi
 s talk is a description of the mathematical structure of the DLN\, especia
 lly the surprising role of minimal cones. The talk will also include some 
 speculation on what the DLN has to tell us about training dynamics in deep
  learning and a description of some common ties\, through Riemannian geome
 try\, between conic programs and deep learning.\n	\n	The talk includes joint
  work with Nadav Cohen (Tel Aviv) and several students at Brown (Lulabel S
 eitz\, Zsolt Veraszto and Tianmin Yu).\n\n \n
DTSTART:20231020T140000Z
DTEND:20231020T150000Z
LOCATION:Room 1214\, Burnside Hall\, CA\, QC\, Montreal\, H3A 0B9\, 805 rue
  Sherbrooke Ouest
SUMMARY:Govind Menon (Brown)
URL:https://www.mcgill.ca/mathstat/channels/event/govind-menon-brown-352098
END:VEVENT
END:VCALENDAR