BEGIN:VCALENDAR
PRODID:-//eluceo/ical//2.0/EN
VERSION:2.0
CALSCALE:GREGORIAN
BEGIN:VEVENT
UID:www.tcs.tifr.res.in/event/523
DTSTAMP:20230914T125928Z
SUMMARY:Randomized Interior Point Methods for Sampling and Optimization
DESCRIPTION:Speaker: Hariharan Narayanan (University of Washington\nDepartm
ent of Statistics and\nDepartment of Mathematics\nPadelford Hall\, Room C-
301\nSeattle\, WA 98105\nUnited States of America)\n\nAbstract: \nAbstract
: Interior point methods are algorithms that optimize convex functions ove
r high dimensional convex sets. From one point of view\, an interior point
method first equips a convex set with a Riemannian metric and then perfor
ms a steepest descent to minimize the objective on the resulting Riemannia
n manifold. We will describe a randomized variant of an interior point met
hod known as ``the affine scaling algorithm" introduced by I.I.Dikin. This
variant corresponds to a natural random walk on the same manifold on whic
h affine scaling would perform steepest descent. We discuss applications t
o sampling and optimization and prove polynomial bounds on the mixing time
of the associated Markov Chain. This talk includes work done in collabo
ration with Ravi Kannan and Alexander Rakhlin.\n
URL:https://www.tcs.tifr.res.in/web/events/523
DTSTART;TZID=Asia/Kolkata:20140805T143000
DTEND;TZID=Asia/Kolkata:20140805T153000
LOCATION:AG69
END:VEVENT
END:VCALENDAR