BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//talks.cam.ac.uk//v3//EN
BEGIN:VTIMEZONE
TZID:Europe/London
BEGIN:DAYLIGHT
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
TZNAME:BST
DTSTART:19700329T010000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
TZNAME:GMT
DTSTART:19701025T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
CATEGORIES:Isaac Newton Institute Seminar Series
SUMMARY:MSG Design of Experiments Seminar Series: The war 
 against bias: experimental design for big data - H
 enry Wynn (London School of Economics)
DTSTART;TZID=Europe/London:20180620T140500
DTEND;TZID=Europe/London:20180620T145500
UID:TALK107143AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/107143
DESCRIPTION:<span>The talk first&nbsp\;reviews&nbsp\; work (by
  others) on optimal  experimental&nbsp\;design for
  &ldquo\;big data&rdquo\;. This ranges from method
 s arising  from the social and medical&nbsp\; scie
 nces\, particularly in causal modelling\,  to rece
 nt work&nbsp\;which tries to extract an optimum de
 sign from a loosely  structured data set of covari
 ates and also&nbsp\;the literature on optimal  des
 ign to guard against bias.&nbsp\;The authors draw&
 nbsp\;on some of&nbsp\;&nbsp\;this work but  take 
 a more game-theoretic approach. The idea is that t
 he causal  modelling operation\, run by an notiona
 l &ldquo\;Alice&rdquo\;\, needs a shield  protecti
 ng against bias built by a notional&nbsp\;&ldquo\;
 Bob&rdquo\;. The two operation can  act&nbsp\; har
 moniously&nbsp\; when the joint operation is a ove
 r&nbsp\;product space  but\, even when not\,&nbsp\
 ; a Nash equilibrium may be achievable\, which  ba
 lances the two objectives. <br></span><br>Joint wo
 rk with Elena Pesce and Eva Riccomagno.
LOCATION:Seminar Room 1\, Newton Institute
CONTACT:INI IT
END:VEVENT
END:VCALENDAR