Title: Bayesian nonparametric methods for conditional independence testing
Authors: Sarah Filippi - Imperial College London (United Kingdom) [presenting]
Abstract: Present Bayesian nonparametric methods for hypothesis testing are presented. In particular, we will focus on methods for quantifying the relative evidence in a dataset in favour of the dependence or independence of two variables conditionally on other variables. The approaches use Polya tree priors on spaces of probability densities, accounting for uncertainty in the form of the underlying distributions in a nonparametric way. The Bayesian perspective provides an inherently symmetric probability measure of conditional dependence or independence, a feature particularly advantageous in causal discovery.