BEGIN:VCALENDAR
PRODID:-//eluceo/ical//2.0/EN
VERSION:2.0
CALSCALE:GREGORIAN
BEGIN:VEVENT
UID:www.tcs.tifr.res.in/event/596
DTSTAMP:20230914T125930Z
SUMMARY:Improved Sample Complexity Estimates for Stochastic Approximation A
lgorithms
DESCRIPTION:Speaker: Gugan Thoppe\n\nAbstract: \nAbstract: The Alexeev's p
erturbation theory gives a method to compare solutions of a perturbed ordi
nary differential equation (ODEs) with that of the unperturbed one. In thi
s talk\, we shall discuss an approach based on this method to obtain sampl
e complexity estimates of stochastic approximation (SA) algorithms. Note t
hat sample complexity refers to the probability that the SA iterates $x_{n
}$ are within an $\\epsilon-$neighbourhood of the equilibrium after lapse
of a certain amount of time\, conditioned on the event that $x_{n_0}$ was
in some bigger neighbourhood of the equilibrium (this is a working paper w
ith Prof. V. Borkar).\n
URL:https://www.tcs.tifr.res.in/web/events/596
DTSTART;TZID=Asia/Kolkata:20150430T140000
DTEND;TZID=Asia/Kolkata:20150430T153000
LOCATION:D-405 (D-Block Seminar Room)
END:VEVENT
END:VCALENDAR