[Abstract]

A fundamental task in various fields of science is to find underlying causal relations and make use of them. Causal relations can be seen if interventions are properly applied; however, it is often difficult to conduct. Hence, it is necessary to discover causal relations by analyzing statistical properties of purely observational data, which is known as causal discovery; however, it is often impossible owing to the non-identifiability. In this talk, I introduce the Gaussian and Poisson directed acyclic graphical (DAG) models for the causal discovery. In addition, I provide how these models can be identifiable without any experiments or interventions, and how the model can be learned in finite sample settings.

[Biography]

Gunwoong Park is currently an assistant professor at the Department of Statistics in Seoul National University, Korea. Previously, he was at the Department of Statistics in University of Seoul, Korea. He received his Ph.D. from the University of Wisconsin and B.S. degree from Seoul National University. His research interests are graphical model learning, high dimensional and robust learning, community detection, and causal reinforcement learning.