BEGIN:VCALENDAR
PRODID:-//eluceo/ical//2.0/EN
VERSION:2.0
CALSCALE:GREGORIAN
BEGIN:VEVENT
UID:www.tcs.tifr.res.in/event/1600
DTSTAMP:20251202T084920Z
SUMMARY:Contextual Generalized Linear Bandits
DESCRIPTION:Speaker: Gaurav Sinha (Microsoft)\n\nAbstract: \nIn this talk\,
  I will present algorithms for the contextual bandit problem with generali
 zed linear rewards. Motivated by practical situations\, I will discuss bat
 ched versions of the problem and develop algorithms that can scale to very
  large action sets while attaining optimal regret with respect to the numb
 er of rounds and the non-linearity of the reward model. Our techniques inc
 lude appropriate optimal designs constructions combined with linear optimi
 zation oracles and action scaling to account for reward nonlinearity. I wi
 ll sketch our proof idea for some of the main results and also present emp
 irical results if time permits.\nShort Bio:I am a Principal Researcher at 
 Microsoft Research\, working in the areas of Reinforcement Learning\, Caus
 al Inference and Learning Theory. I received my Ph.D. in Mathematics from
  the California Institute of Technology in 2016\, where I was advised by P
 rof. Eric Rains and my Integrated M.Sc. in Mathematics and Scientific Comp
 uting from Indian Institute of Technology (IIT) Kanpur in 2011. My primary
  research interest lies in sequential decision-making\, particularly as ap
 plied to real world applications such as online advertising and recommenda
 tion systems. I focus on scenarios where decision-makers have access to bo
 th observational and interventional data. While observational data is abun
 dant and inexpensive\, interventional data\, though costly\, provides more
  granular insights into the impact of specific decisions. My research aims
  to develop algorithms that optimally balance the use of these data types\
 , ensuring both theoretically sound and practically viable decision-making
  over extended periods. To achieve this\, I employ a diverse toolkit encom
 passing causal inference\, nonlinear and stochastic optimization\, optimal
  experimental design etc.\n
URL:https://www.tcs.tifr.res.in/web/events/1600
DTSTART;TZID=Asia/Kolkata:20251202T160000
DTEND;TZID=Asia/Kolkata:20251202T170000
LOCATION:A-201 (STCS Seminar Room)
END:VEVENT
END:VCALENDAR