The 2016 ACM SIGKDD Workshop on Causal Discovery

                                     Held in conjunction with KDD'16                         

August 14, 2016, San Francisco, California

Invited talk 

Title: Foundations of Causal Discovery  (Slides)

Presenter: Professor Frederick Eberhardt

Abstract: The now widely used theory of causal graphical models considers causal relations among a set of statistical variables. The causal relations are represented in terms of a directed graph among the set of variables, and the task of causal discovery is to identify this causal structure on the basis of the probability distribution generated by the variables in the graph. I will provide an introduction and overview of some of the methods for causal discovery and present known identifiability results with a particular focus on the assumptions they depend on.

Biography: Frederick Eberhardt is Professor of Philosophy at Caltech. His research focuses on the development of causal discovery algorithms, the foundations of causal inference and the relation between causality and probability. Apart from research in methodology, he is also interested in how humans and animals actually learn causal relations. Before joining Caltech, he was Assistant Professor in the philosophy-neuroscience-psychology program at Washington University in St Louis. He did his PhD in philosophy at Carnegie Mellon.