BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Identification of causal effects - Nevena Lazic
DTSTART:20121025T133000Z
DTEND:20121025T150000Z
UID:TALK41262@talks.cam.ac.uk
CONTACT:Konstantina Palla
DESCRIPTION:Establishing cause-effect relationships from a combination of 
 data and assumptions is a fundamental part of empirical science. \nGraphic
 al models provide a useful framework for representing assumptions about th
 e world and formalizing causal inference. \nIn this talk\, I will first de
 scribe a complete algorithm by Tian & Pearl for determining whether a caus
 al effect is identifiable from \nobservational data for a given graphical 
 model. I will then discuss the relationship between identifiable effects a
 nd recursive \nfactorization of the observational distribution\, with pote
 ntial implications for computationally efficient inference. 
LOCATION:Engineering Department\, CBL Room BE-438
END:VEVENT
END:VCALENDAR