BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:First-order methods for the impatient - support identification in 
 finite time with Frank/Wolfe-variants
DTSTART:20241122T090000Z
DTEND:20241122T094000Z
DTSTAMP:20260610T120400Z
UID:indico-event-13450@indico.math.cnrs.fr
DESCRIPTION:Speakers: Abdel Lisser\n\nWe study active set identification r
 esults for the away-step Frank-Wolfe algorithm in different settings. We f
 irst prove a local identification property that we apply\, in combination 
 with a convergence hypothesis\, to get an active set identification result
 . We then prove\, in the nonconvex case\, a novel $O(1/\\sqrt{k})$ converg
 ence rate result and active set identification for different step sizes (u
 nder suitable assumptions on the set of stationary points). By exploiting 
 those results\, we also give explicit active set complexity bounds for bot
 h strongly convex and nonconvex objectives. While we initially consider th
 e probability simplex as feasible set\, time permitting we show how to ada
 pt some of our results to generic polytopes. A particular case with intere
 sting applications covers projection-free methods on product domains.\n\nh
 ttps://indico.math.cnrs.fr/event/13450/
LOCATION:Salle de conférence (XLIM)
URL:https://indico.math.cnrs.fr/event/13450/
END:VEVENT
END:VCALENDAR