BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Markov categories and the inflation technique for latent-variable 
 causal inference - Tobias Fritz (Leopold-Franzens Universtät Innsbruck)
DTSTART:20260123T094500Z
DTEND:20260123T103000Z
UID:TALK241840@talks.cam.ac.uk
DESCRIPTION:Markov categories have recently gained prominence as a higher-
 level language for probability and statistics\, and for causal inference i
 n particular. This language&nbsp\;facilitates a treatment of continuous va
 riables and general measure-theoretic probability with the same ease as fo
 r discrete variables and facilitates clear and general proofs of many qual
 itative results (such as the d-separation criterion or certain 0/1-laws). 
 In this talk\, I will give a gentle introduction to Markov categories\, an
 d subsequently illustrate the formalism by presenting the inflation techni
 que&nbsp\;in terms of Markov categories. The inflation technique is a hier
 archy of tests which are necessary and (in the limit) sufficient condition
 s to decide the compatibility of a given probability distribution with a g
 iven causal structure with latent variables.
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR