Timezone: »

Conditional Independence Testing using Generative Adversarial Networks
Alexis Bellot · Mihaela van der Schaar

Tue Dec 10 10:40 AM -- 10:45 AM (PST) @ West Ballroom A + B

We consider the hypothesis testing problem of detecting conditional dependence, with a focus on high-dimensional feature spaces. Our contribution is a new test statistic based on samples from a generative adversarial network designed to approximate directly a conditional distribution that encodes the null hypothesis, in a manner that maximizes power (the rate of true negatives). We show that such an approach requires only that density approximation be viable in order to ensure that we control type I error (the rate of false positives); in particular, no assumptions need to be made on the form of the distributions or feature dependencies. Using synthetic simulations with high-dimensional data we demonstrate significant gains in power over competing methods. In addition, we illustrate the use of our test to discover causal markers of disease in genetic data.

Author Information

Alexis Bellot (University of Cambridge / Alan Turing Institute)
Mihaela van der Schaar (University of Cambridge, Alan Turing Institute and UCLA)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors