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:Statistics
SUMMARY:What's the right complexity measure for inferring 
 causal relations? - Dominik Janzing (Max Planck In
 stitute Tuebingen)
DTSTART;TZID=Europe/London:20100226T153000
DTEND;TZID=Europe/London:20100226T163000
UID:TALK22581AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/22581
DESCRIPTION:If X causes Y for two random variables X and Y\, w
 e expect that the\nfactorization of P(X\,Y) into P
 (X)P(Y|X) is simpler than the *non-causal*\nfactor
 ization P(Y)P(X|Y). This is because P(Y|X) describ
 es the causal\nmechanism while P(X|Y) is "only a m
 athematical expression".\n\nDiscussions have shown
  that a lot of researchers agree on this\nintuitio
 n. Since we would like to use this principle for i
 nferring\ncausal directions\, we are left with two
  problems:\n\n(1) what does "simple" mean?\n\n(2) 
 is there any deeper justification for this princip
 le?\n\nOur answer to question (2) is a clear "yes"
  if complexity vs simplicity\nis measured in terms
  of Kolmogorov complexity: I will present a theory
 \nof causal inference that generalizes the framewo
 rk of Bayesian networks\nto *algorithmic* instead 
 of *statistical* conditional dependences. I\nwill 
 show that our theory implies the above inference p
 rinciple.\nHowever\, since Kolmogorov complexity i
 s uncomputable we still need\ncomplexity measures 
 that are appropriate for practical implementations
 .\n\nI will present some first small steps towards
  this challenging goal.\n\n\n\n\nhttp://www.kyb.mp
 g.de/~janzing
LOCATION:MR12\, CMS\, Wilberforce Road\, Cambridge\, CB3 0W
 B
CONTACT:
END:VEVENT
END:VCALENDAR