BEGIN:VCALENDAR
PRODID:-//eluceo/ical//2.0/EN
VERSION:2.0
CALSCALE:GREGORIAN
BEGIN:VEVENT
UID:www.tcs.tifr.res.in/event/1760
DTSTAMP:20260807T050059Z
SUMMARY:Wealth\, Duality\, and Regret: The Coin Betting Framework for Onlin
 e Linear Optimization
DESCRIPTION:Speaker: Aakash Ghosh (TIFR)\n\nAbstract: \nConsider a gambler 
 who repeatedly bets a fraction of her wealth on the outcome of an adversar
 ially chosen coin\, and suppose we ask only that her wealth grow at a cert
 ain rate relative to that of the best fixed betting fraction in hindsight.
  This talk develops the observation\, due to Orabona and Pál\, that this 
 elementary game is not merely an analogy for online learning but is equiva
 lent to it: a betting algorithm with a suitable wealth guarantee mechanica
 lly yields parameter-free algorithms for online linear optimization over H
 ilbert spaces and for learning with expert advice\, with regret bounds tha
 t adapt to the unknown comparator without oracle tuning of a learning rate
 .\n
URL:https://www.tcs.tifr.res.in/web/events/1760
DTSTART;TZID=Asia/Kolkata:20260807T160000
DTEND;TZID=Asia/Kolkata:20260807T170000
LOCATION:A-201 (STCS Seminar Room)
END:VEVENT
END:VCALENDAR
