Causal Discovery Machine Learning: Research Applications for Psychology, Neuroscience, and Psychiatry
This five-day summer school is an introduction to the applied use of causal discovery machine learning for graduate students, postdocs, and faculty who are conducting research in psychology, neuroscience, and psychiatry. Path analysis methods (e.g., structural equation modeling) require a priori specification of a causal model, but in most cases the true causal model underlying relationships in observed data is unknown. Causal discovery analysis instead uses a purely data-driven approach to search over the parameter space of potential causal models that could give rise to the observed relationships in data and return the best-fitting model.
Over the course of this summer school, we will first cover the assumptions, limitations, and uses of various causal discovery analysis methods. We will then move on to discussing more advanced topics such as applying causal discovery analysis to longitudinal data, neural connectivity, and discovery and estimation of latent variables. Finally, students will coordinate with the summer school organizers to develop an individual project based on their own research interests, and will receive continued support for that project following conclusion of the summer school.
Additional information is available on The Flyer!
Application Deadline: May 15, 2021
Date of the Course: July 5-9, 2021
How to Apply: Interested applicants should email a CV and a one-page (max) statement of interest describing a) your current research and b) potential applications of causal discovery analysis in your research to email@example.com.
Organizers: Eric Rawls, Ph.D., Erich Kummerfeld, Ph.D., Sisi Ma, Ph.D.
This summer school is supported by the William K. and Katherine W. Estes Fund, which is jointly overseen by the Association for Psychological Science and the Psychonomics Society.