BEGIN:VCALENDAR
VERSION:2.0
PRODID:researchseminars.org
CALSCALE:GREGORIAN
X-WR-CALNAME:researchseminars.org
BEGIN:VEVENT
SUMMARY:Mengdi Wang (Princeton University)
DTSTART:20200526T190000Z
DTEND:20200526T200000Z
DTSTAMP:20260404T110643Z
UID:OptMLStat/1
DESCRIPTION:Title: <a href="https://stable.researchseminars.org/talk/OptML
 Stat/1/">Statistical complexity of reinforcement learning</a>\nby Mengdi W
 ang (Princeton University) as part of Online Seminar of Mathematical Found
 ations of Data Science\n\nAbstract: TBA\n
LOCATION:https://stable.researchseminars.org/talk/OptMLStat/1/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Yinyu Ye (Stanford University)
DTSTART:20200602T190000Z
DTEND:20200602T200000Z
DTSTAMP:20260404T110643Z
UID:OptMLStat/2
DESCRIPTION:Title: <a href="https://stable.researchseminars.org/talk/OptML
 Stat/2/">Distributionally robust optimization\, online linear programming 
 and markets for public-good allocations</a>\nby Yinyu Ye (Stanford Univers
 ity) as part of Online Seminar of Mathematical Foundations of Data Science
 \n\nAbstract: TBA\n
LOCATION:https://stable.researchseminars.org/talk/OptMLStat/2/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Robert M. Freund (MIT)
DTSTART:20200609T190000Z
DTEND:20200609T200000Z
DTSTAMP:20260404T110643Z
UID:OptMLStat/3
DESCRIPTION:Title: <a href="https://stable.researchseminars.org/talk/OptML
 Stat/3/">From stochastic Frank-Wolfe to the ellipsoid method: Recent progr
 ess on practical optimization in data science (the Frank-Wolfe Method) and
  theoretical optimization (the ellipsoid method)</a>\nby Robert M. Freund 
 (MIT) as part of Online Seminar of Mathematical Foundations of Data Scienc
 e\n\nAbstract: TBA\n
LOCATION:https://stable.researchseminars.org/talk/OptMLStat/3/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Sham Kakade (University of Washington\, Seattle)
DTSTART:20200616T190000Z
DTEND:20200616T200000Z
DTSTAMP:20260404T110643Z
UID:OptMLStat/4
DESCRIPTION:Title: <a href="https://stable.researchseminars.org/talk/OptML
 Stat/4/">Representation\, Modeling\, and Gradient Based Optimization in Re
 inforcement Learning</a>\nby Sham Kakade (University of Washington\, Seatt
 le) as part of Online Seminar of Mathematical Foundations of Data Science\
 n\n\nAbstract\nReinforcement learning is now the dominant paradigm for how
  an agent learns to interact with the world. The approach has lead to succ
 esses ranging across numerous domains\, including game playing and robotic
 s\, and it holds much promise in new domains\, from self driving cars to i
 nteractive medical applications. Some of the central challenges are:\n\n- 
 Representational learning: does having a good representation of the enviro
 nment permit efficient reinforcement learning?\n\n- Modeling: should we ex
 plicitly build a model of our environment or\, alternatively\, should we d
 irectly learn how to act?\n\n- Optimization: in practice\, deployed algori
 thms often use local search heuristics. Can we provably understand  when t
 hese approaches are effective and provide faster and more robust alternati
 ves?\n\nThis talk will survey a number of results on these basic questions
 . Throughout\, we will  highlight the interplay of theory\, algorithm desi
 gn\, and practice.\n
LOCATION:https://stable.researchseminars.org/talk/OptMLStat/4/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Robert Nowak (University of Wisconsin\, Madison)
DTSTART:20200623T190000Z
DTEND:20200623T200000Z
DTSTAMP:20260404T110643Z
UID:OptMLStat/5
DESCRIPTION:by Robert Nowak (University of Wisconsin\, Madison) as part of
  Online Seminar of Mathematical Foundations of Data Science\n\nAbstract: T
 BA\n
LOCATION:https://stable.researchseminars.org/talk/OptMLStat/5/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Alex Shapiro (Georgia Institute of Technology)
DTSTART:20200702T190000Z
DTEND:20200702T200000Z
DTSTAMP:20260404T110643Z
UID:OptMLStat/6
DESCRIPTION:by Alex Shapiro (Georgia Institute of Technology) as part of O
 nline Seminar of Mathematical Foundations of Data Science\n\nAbstract: TBA
 \n
LOCATION:https://stable.researchseminars.org/talk/OptMLStat/6/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Adrian S. Lewis (Cornell University)
DTSTART:20200707T190000Z
DTEND:20200707T200000Z
DTSTAMP:20260404T110643Z
UID:OptMLStat/7
DESCRIPTION:Title: <a href="https://stable.researchseminars.org/talk/OptML
 Stat/7/">Smoothness in nonsmooth optimization</a>\nby Adrian S. Lewis (Cor
 nell University) as part of Online Seminar of Mathematical Foundations of 
 Data Science\n\nAbstract: TBA\n
LOCATION:https://stable.researchseminars.org/talk/OptMLStat/7/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Michael I. Jordan (UC Berkeley)
DTSTART:20200714T190000Z
DTEND:20200714T200000Z
DTSTAMP:20260404T110643Z
UID:OptMLStat/8
DESCRIPTION:Title: <a href="https://stable.researchseminars.org/talk/OptML
 Stat/8/">Optimization with momentum: dynamical\, variational\, and symplec
 tic perspectives</a>\nby Michael I. Jordan (UC Berkeley) as part of Online
  Seminar of Mathematical Foundations of Data Science\n\nAbstract: TBA\n
LOCATION:https://stable.researchseminars.org/talk/OptMLStat/8/
END:VEVENT
END:VCALENDAR