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:Causal Discovery in Network Data - Elena Zheleva (
 University of Illinois Chicago)
DTSTART;TZID=Europe/London:20260306T103000
DTEND;TZID=Europe/London:20260306T111500
UID:TALK244429AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/244429
DESCRIPTION:Most existing causal discovery algorithms rely on 
 the assumptions that data are independent and iden
 tically distributed (i.i.d.). However\, many real-
 world domains\, such as biological and social netw
 orks\, violate the i.i.d. assumption and consist o
 f interacting entities whose attributes exhibit co
 mplex relational and causal dependencies\, breakin
 g the SUTVA assumption and leading to interference
 . To address these challenges and to facilitate ca
 usal reasoning in network settings\, I will presen
 t two recent contributions that develop graphical 
 models and algorithms for causal discovery in netw
 ork data in the presence of cycles and latent vari
 ables. The first contribution introduces relationa
 l acyclification\, an operation specifically desig
 ned for cyclic relational causal models that enabl
 es formal analysis of identifiability in cyclic re
 lational structures. Under the assumptions of rela
 tional acyclification and &sigma\;-faithfulness\, 
 we establish that the Relational Causal Discovery 
 (RCD) algorithm (Maier et al.\, 2013) is sound and
  complete for models containing cyclic dependencie
 s. The second contribution presents RelFCI\, a cau
 sal discovery algorithm that is sound and complete
  for relational data subject to latent confounding
 . In this work\, we further derive soundness and c
 ompleteness guarantees for relational d-separation
  in the presence of latent variables\, thereby ext
 ending causal discovery theory to a broader class 
 of relational systems.
LOCATION:Seminar Room 1\, Newton Institute
CONTACT:
END:VEVENT
END:VCALENDAR