Identifying Intended Effects with Causal Models
Résumé
The aim of this work is to extend the framework of causal inference, in particular as
it has been developed by Pearl (2009), in order to model actions and their ends.
It introduces means-ends relationships as a way to model actions based on a
causal model and of its implied correlations observable in data; it defines
teleological confounding and introduces interference as a way to control for it.
| Origine | Fichiers produits par l'(les) auteur(s) |
|---|