BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4//
BEGIN:VEVENT
UID:20251110T053338EST-8967uLpXlc@132.216.98.100
DTSTAMP:20251110T103338Z
DESCRIPTION:Title: Differentiating Nonsmooth Solutions to Parametric Monoto
 ne Inclusion Problems\n\nAbstract :\n\nUnderstanding the differentiability
  and regularity of the solution to a monotone inclusion problem is an impo
 rtant question with consequences for convex optimization\, deep learning w
 ith implicit layers\, and beyond. Past attempts at answering this question
  have been made either under very restrictive assumptions that ensure the 
 solution is continuously differentiable or using mathematical tools that a
 re incompatible with automatic differentiation. In this talk\, we discuss 
 how to leverage path differentiability and a recent result on nonsmooth im
 plicit differentiation calculus to give sufficient conditions ensuring tha
 t the solution to a monotone inclusion problem will be path differentiable
  and provide formulas for computing its generalized gradient. Our approach
  is fully compatible with automatic differentiation and comes with assumpt
 ions which are easy to check\, roughly speaking: semialgebraicity and stro
 ng monotonicity. We illustrate the scope of our results by considering thr
 ee fundamental composite problem settings: strongly convex problems\, dual
  solutions to convex minimization problems and primal-dual solutions to mi
 n-max problems.\n\n \n
DTSTART:20240205T210000Z
DTEND:20240205T220000Z
LOCATION:Room 1104\, Burnside Hall\, CA\, QC\, Montreal\, H3A 0B9\, 805 rue
  Sherbrooke Ouest
SUMMARY:Antonio Silveti-Falls (Universite Paris-Saclay)
URL:/mathstat/channels/event/antonio-silveti-falls-uni
 versite-paris-saclay-355161
END:VEVENT
END:VCALENDAR