Conditional Independence Testing using Generative Adversarial Networks

2019/07/09 by Alexis Bellot, Mihaela van der Schaar, Bellot, Alexis +1 · 6 citations
Computer Science · Mathematics · #FOS: Computer and information sciences #Image and Object Detection Techniques #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Machine Learning and Algorithms #Machine Learning and Data Classification #cs.LG #stat.ML

paper · pdf · doi:10.48550/arxiv.1907.04068

Updated version published at NeurIPS 2019

openalex publication_date 2019/07/09 · arxiv created 2019/12/18 · arxiv updated 2019/12/20 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

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.

Citations

• Kernel-based Conditional Independence Test and Application in Causal Discovery
• Systematic identification of genomic markers of drug sensitivity in cancer cells
• Energy-based Generative Adversarial Network
• Conditional independence testing based on a nearest-neighbor estimator of conditional mutual information

Cited by

• Toward Scalable and Valid Conditional Independence Testing with Spectral Representations
• Efficient Ensemble Conditional Independence Test Framework for Causal Discovery
• A Conditional Distribution Equality Testing Framework using Deep Generative Learning
• AICO: Feature Significance Tests for Supervised Learning
• A Fast Kernel-based Conditional Independence test with Application to Causal Discovery
• PAIR-CI: Calibrated Conditional Independence Testing for Causal Discovery with Incomplete Data

Related

• Double Generative Adversarial Networks for Conditional Independence Testing
• Amortized Conditional Independence Testing
• Doubly Robust Conditional Independence Testing with Generative Neural Networks
• Mimic and Classify : A meta-algorithm for Conditional Independence Testing
• Score-based Generative Modeling for Conditional Independence Testing