Research map / Probabilistic and scientific modeling / Causal discovery
A Distribution Free Conditional Independence Test with Applications to Causal Discovery
Zhanrui Cai, Runze Li, Yaowu Zhang
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 CIT instead of conditional distance correlation (CDC) | conditional independence testing · M3, n=50, alpha=0.05 | empirical power: 0.338 -> 0.572 | Improved |
| use CIT instead of CMI | conditional independence testing · M3, n=100, alpha=0.05 | empirical power: 0.070 -> 0.960 | Improved |
| use CIT instead of KCI | conditional independence testing · M6 heavy-tailed, n=50, alpha=0.05 | empirical power: 0.010 -> 0.888 | Improved |
| use CIT instead of CDC with 2-dimensional Z | conditional independence testing · M10, n=50, alpha=0.05 | empirical power: 0.054 -> 0.686 | Improved |
| use CIT instead of KCI with 2-dimensional Z | conditional independence testing · M11, n=50, alpha=0.05 | empirical power: 0.440 vs 0.662 | Hurt |
| use CIT instead of CDC with multivariate X, Y, Z | conditional independence testing · M14, n=50, alpha=0.05 | empirical power: 0.022 -> 1.000 | Improved |
| vary kernel bandwidth multiplier c from 0.5 to 1.5 (ablation) | CIT · M2-M6, n=100, alpha=0.05 | empirical power: 0.957 to 0.98 for M3; 1.000 for M2 | No clear effect |
| use CIT instead of CMI and KCI when conditioning set is large | conditional independence testing · M7-M18 | empirical power | Improved |
| use CIT in PC algorithm instead of partial correlation on log-transformed data | PC algorithm causal discovery · diabetes data, n=392 | graph stability | Improved |
Closest papers by meaning
- Causal Discovery via Conditional Independence Testing with Proxy Variables (ICML 2024 · similarity 0.90)
- PAIR-CI: Calibrated Conditional Independence Testing for Causal Discovery with Incomplete Data (NeurIPS 2026 · similarity 0.89)
- On the sample complexity of conditional independence testing with Von Mises estimator with application to causal discovery (ICML 2024 · similarity 0.88)
- A Conditional Independence Test in the Presence of Discretization (ICLR 2025 · similarity 0.88)
- Extracting Rare Dependence Patterns via Adaptive Sample Reweighting (ICML 2025 · similarity 0.87)
- A Sample Efficient Conditional Independence Test in the Presence of Discretization (ICML 2025 · similarity 0.87)
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.
