Research map / Image generation / GANs
Conditional Independence Testing using Generative Adversarial Networks
Alexis Bellot, Mihaela van der Schaar
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use GCIT instead of CRT or kernel-based tests | Conditional independence testing · Synthetic data, multivariate Gaussian | power | No clear effect |
| Use GCIT instead of CRT or kernel-based tests | Conditional independence testing · Synthetic data, multivariate Laplace and arbitrary distributions | power | Improved |
| Add the information network to the GCIT generator objective | GCIT · Synthetic data | power | Improved |
| Increase the information-network weight λ (ablation) | GCIT · Synthetic data, multivariate Gaussian | power | Improved |
| Increase the information-network weight λ (ablation) | GCIT · Synthetic data under the null | type I error | Hurt |
| Train a GCIT model rather than using CCIT | Conditional independence testing · Synthetic data, >1000 samples or 500 Z dimensions | runtime: over 600s versus 60s | Improved |
| Use GCIT to test genetic mutation–drug response dependence | GCIT · CCLE PLX4720 | p-value | Mixed |
Closest papers by meaning
- Double Generative Adversarial Networks for Conditional Independence Testing (JMLR 2021 · similarity 0.93)
- Testing Conditional Mean Independence Using Generative Neural Networks (ICML 2025 · similarity 0.88)
- A Likelihood Based Approach to Distribution Regression Using Conditional Deep Generative Models (ICML 2025 · similarity 0.86)
- Conditional Diffusion Models Based Conditional Independence Testing (AAAI 2025 · similarity 0.86)
- On the Hardness of Conditional Independence Testing In Practice (NeurIPS 2025 · similarity 0.85)
- Conditional Generative Models are Sufficient to Sample from Any Causal Effect Estimand (NeurIPS 2024 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
