BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//ZContent.net//ZapCalLib 1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VEVENT
DTSTART;TZID=Atlantic/Canary:20231024T123000
DTEND;TZID=Atlantic/Canary:20231024T133000
UID:iactalks-1725
X-WR-CALNAME: IAC Talks: Open Astronomy Seminars
X-ORIGINAL-URL: /iactalks/Talks/view/1725
CREATED:2023-10-24T12:30:00+01:00
X-WR-CALDESC: IAC Talks upcomming talks
SUMMARY:Beyond Predictions: Causal Models in Artificial Intelligence
DESCRIPTION:Beyond Predictions: Causal Models in Artificial Intelligence\nL
 . Enrique Sucar\n\nCurrent intelligent systems mainly make "predictions". 
 That is, given an input, they estimate the most probable value of the outp
 ut. These systems have many limitations, they can be easily confused when 
 presented with a different case from their training set and they cannot ex
 plain how they arrive to a certain result. Causal models are an alternativ
 e to extend the capabilities of current systems; explain the reasons for c
 ertain decisions, predict the effect of interventions and imagine alternat
 ive situations. In this talk, I will present an introduction to causal mod
 els, in particular to causal graphical models. We will see how we can make
  inferences based on these models: predictions and counterfactuals; as wel
 l as learning causal models from data. I will illustrate the application o
 f causal models in various domains: estimation of effective connectivity i
 n the brain, causal modeling of COVID-19, and incorporating causal models 
 in reinforcement learning and its application in robotics. Finally, I will
  discuss some potential applications of causal modeling in astrophysics.
END:VEVENT
END:VCALENDAR