Conditional Independence Testing using Generative Adversarial Networks

Abstract

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.

arXiv commentUpdated version published at NeurIPS 2019

Cited by in corpus (5)

• Fast and Powerful Conditional Randomization Testing via Distillation
• Data-Driven Representations for Testing Independence: Modeling, Analysis and Connection with Mutual Information Estimation
• Double Generative Adversarial Networks for Conditional Independence Testing
• Deconfounded Score Method: Scoring DAGs with Dense Unobserved Confounding
• Kernel Hypothesis Testing with Set-valued Data