BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4//
BEGIN:VEVENT
UID:20260913T091426EDT-4666SDDwzd@132.216.98.100
DTSTAMP:20260913T131426Z
DESCRIPTION:Title: Convergence rates for diffusions-based sampling and opti
 mization methods\n\nAbstract: An Euler discretization of the Langevin diff
 usion is known to converge to the global minimizers of certain convex and 
 non-convex optimization problems. We show that this property holds for any
  suitably smooth diffusion and that different diffusions are suitable for 
 optimizing different classes of convex and non-convex functions. This allo
 ws us to design diffusions suitable for globally optimizing convex and non
 -convex functions not covered by the existing Langevin theory. Our non-asy
 mptotic analysis delivers computable optimization and integration error bo
 unds based on easily accessed properties of the objective and chosen diffu
 sion. Central to our approach are new explicit Stein factor bounds on the 
 solutions of Poisson equations. We complement these results with improved 
 optimization guarantees for targets other than the standard Gibbs measure.
 \n
DTSTART:20191129T203000Z
DTEND:20191129T213000Z
LOCATION:Room 1205\, Burnside Hall\, CA\, QC\, Montreal\, H3A 0B9\, 805 rue
  Sherbrooke Ouest
SUMMARY:Murat A. Erdoglu (University of Toronto)
URL:https://www.mcgill.ca/mathstat/channels/event/murat-erdoglu-university-
 toronto-302916
END:VEVENT
END:VCALENDAR