BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:A novel (pre-)metric for causal graphs - Jonas Peters\, ETH Züric
 h
DTSTART:20150130T143000Z
DTEND:20150130T153000Z
UID:TALK56939@talks.cam.ac.uk
CONTACT:20082
DESCRIPTION:Causal discovery methods estimate a system's underlying causal
  graph from observational data. But even if the true structure is known\, 
 it is not clear how to evaluate the performance of this estimator. In many
  cases\, we use the causal graph to infer the system's behaviour under spe
 cific interventions\, i.e. manipulations of some of the variables. Differe
 nt graphs may then result in different predicted behaviours. The structura
 l intervention distance (SID) quantifies these differences and thereby def
 ines a (pre-)metric between graphs. We briefly discuss recent results on i
 dentification of causal structure and show that they are closely related t
 o the concept of SID.\nThis talk does not require any prior knowledge abou
 t causal concepts.
LOCATION:MR12\,  Centre for Mathematical Sciences\, Wilberforce Road\, Cam
 bridge
END:VEVENT
END:VCALENDAR