BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Towards Neuro-Causality: Relating Graph Neural Networks to Structu
 ral Causal Models - Matej Zecevic\, TU Darmstadt
DTSTART:20210924T130000Z
DTEND:20210924T140000Z
UID:TALK162220@talks.cam.ac.uk
CONTACT:Chaochao Lu
DESCRIPTION:Xia\, Lee\, Bengio and Bareinboim recently formalized the Caus
 al-Neural Connection in spirit of previously existing work (e.g. Kocaoglu 
 et al. 2017\, Ke et al. 2020). This talk will start with an introduction t
 o this arguably new research direction of interest: Neuro-Causality. Think
 ing of pure Causality as formalized by Judea Pearl in his seminal work\, i
 t can be described in terms of a Structural Causal Model (SCM) that carrie
 s information on the variables of interest and their mechanistic relations
 . For most processes of interest the underlying SCM will only be partially
  observable\, thus causal inference tries to leverage any exposed informat
 ion. Most recently\, Zečević\, Dhami\, Veličković\, and Kersting consi
 dered the special network type known as Graph Neural Networks (GNN)\, whic
 h act as universal approximators on structured input\, for causal learning
  - thereby suggesting a tighter integration with SCM. For said work\, star
 ting from first principles the talk will examine key theoretical results. 
 Finally\, the talk will conclude with a perspective on interesting future 
 research directions for neuro-causality.
LOCATION:https://cuhk-edu-cn.zoom.us/j/91529611509?pwd=WFBqeFdVc3J1cE9rV2N
 aMXJtM2RQQT09
END:VEVENT
END:VCALENDAR