BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:(CANCELLED) Stochastic Causal Programming for Bounding Treatment E
 ffects - Ricardo Silva (UCL)
DTSTART:20220617T130000Z
DTEND:20220617T140000Z
UID:TALK173327@talks.cam.ac.uk
CONTACT:Qingyuan Zhao
DESCRIPTION:Causal effect estimation is important for numerous tasks in th
 e natural and social sciences. However\, identifying effects is impossible
  from observational data without making strong\, often untestable assumpti
 ons. We consider algorithms for the partial identification problem\, bound
 ing treatment effects from multivariate\, continuous treatments over multi
 ple possible causal models when unmeasured confounding makes identificatio
 n impossible. We consider a framework where observable evidence is matched
  to the implications of constraints encoded in a causal model by norm-base
 d criteria. This generalizes classical approaches based purely on generati
 ve models. Casting causal effects as objective functions in a constrained 
 optimization problem\, we combine flexible learning algorithms with Monte 
 Carlo methods to implement a family of solutions under the name of stochas
 tic causal programming. In particular\, we present ways by which such cons
 trained optimization problems can be parameterized without likelihood func
 tions for the causal or the observed data model\, reducing the computation
 al and statistical complexity of the task.\n \nJoint work with Kirtan Padh
 \, Jakob Zeitler\, David Watson\, Matt Kusner and Niki Kilbertus
LOCATION:MR12\, Centre for Mathematical Sciences
END:VEVENT
END:VCALENDAR