BEGIN:VCALENDAR
VERSION:2.0
CALSCALE:GREGORIAN
PRODID:iCalendar-Ruby
BEGIN:VEVENT
CATEGORIES:Conferences/Seminars/Lectures
DESCRIPTION:Moshen Bayati\nAssociate Professor\nStanford University\n\nAbs:
  A central problem in personalized decision-making is to learn decision out
 comes as functions of individual-specific covariates (contexts). Current li
 terature on this topic focuses on algorithms that balance an exploration-ex
 ploitation tradeoff\, to ensure sufficient rate of learning while optimizin
 g for some objective. However\, exploration may be undesirable for highly s
 ensitive individuals (e.g.\, patients in clinical treatment planning). In t
 his talk\, we first introduce an algorithm that leverages free-exploration 
 from the covariates and achieves rate optimal objective. Moreover\, we show
  empirically that our algorithm significantly reduces exploration\, compare
 d to existing benchmarks. Next\, we focus on settings when past data on dec
 ision outcomes is available or when the number of decisions is large. Motiv
 ated by literature on low-rank matrix estimation\, we design algorithms tha
 t avoid unnecessary exploration by targeting the learning towards shared si
 milarities among decisions or patients.\n\n\nBio: Mohsen received is associ
 ate professor of Operations\, Information\, and Technology at Stanford Univ
 ersity Graduate School of Business. Prior to joining Stanford faculty in 20
 11\, he was postdoc in Stanford University and Microsoft Research. MohsenÕs
  research is on healthcare management\, statistical inference via graphical
  models\, and personalized decision-making. His research as received the IN
 FORMS Healthcare Applications Society best paper (Pierskalla) award in 2014
  and in 2016\, INFORMS Applied Probability Society best paper award in 2015
 \, and National Science Foundation CAREER award.
DTEND:20181115T221500Z
DTSTAMP:20260314T092119Z
DTSTART:20181115T211500Z
LOCATION:E51-335\, E51-335
SEQUENCE:0
SUMMARY:Reducing Exploration in Personalized Decision-Making
UID:tag:localist.com\,2008:EventInstance_3867408
URL:https://calendar.mit.edu/event/a_talk_on_personalized_decision_making
END:VEVENT
END:VCALENDAR