Projects per year
Abstract
Computing the effects of interventions from observational data is an important task encountered in many datadriven sciences. The problem is addressed by identifying the postinterventional distribution with an expression that involves only quantities estimable from the preinterventional distribution over observed variables, given some knowledge about the causal structure. In this work, we relax the requirement of having a fully specified causal structure and study the identifiability of effects with a singleton intervention (X), supposing that the structure is known only up to an equivalence class of causal diagrams, which is the output of standard structural learning algorithms (e.g., FCI). We derive a necessary and sufficient graphical criterion for the identifiability of the effect of X on all observed variables. We further establish a sufficient graphical criterion to identify the effect of X on a subset of the observed variables, and prove that it is strictly more powerful than the current stateoftheart result on this problem.
Original language  English 

Title of host publication  Proceedings of the TwentySeventh International Joint Conference on Artificial Intelligence (IJCAI18) 
Editors  Jérôme Lang LANG 
Publisher  International Joint Conferences on Artificial Intelligence 
Pages  50245030 
ISBN (Print)  9780999241127 
DOIs  
Publication status  Published  Jul 2018 
Fingerprint
Dive into the research topics of 'A Graphical Criterion for Effect Identification in Equivalence Classes of Causal Diagrams'. Together they form a unique fingerprint.Projects
 1 Finished

Causation, Decision, and Imprecise Probabilities
ZHANG, J. (PI) & SEIDENFELD, T. (CoI)
Research Grants Council (HKSAR)
1/01/16 → 31/12/17
Project: Grant Research