Research map / Learning theory and algorithms / Kernels and regression theory
Measuring Dependence Powerfully and Equitably
Yakir A. Reshef, David N. Reshef, Hilary K. Finucane, Pardis C. Sabeti, Michael Mitzenmacher
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 EquicharClump to compute MICe instead of Approx-MIC | MIC estimation · Runtime, n=500 to n=10,000 | speedup: 30-fold speedup at n = 500 to over a 350-fold speedup at n = 10,000 | Improved |
| use TICe instead of distance correlation | independence testing · Simon-Tibshirani relationships (n=500) | power: uniformly outperforms on 5 of the 8 relationship types | Improved |
| use TICe instead of distance correlation | independence testing · Simon-Tibshirani relationships (n=500) | power: comparable on the other three relationship types | No clear effect |
| sum over the characteristic matrix (TICe) instead of maximizing (MICe) (ablation) | independence testing · Simon-Tibshirani relationships (n=500) | power | Improved |
| use TICe instead of original MIC with default parameters | independence testing · Simon-Tibshirani relationships (n=500) | power | Improved |
| use MICe instead of MIC | equitability w.r.t. R2 · noisy functional relationships | equitability | Improved |
| increase the exponent alpha in MICe | MICe · low-signal regimes | estimation error | Hurt |
| decrease the exponent alpha in MICe | MICe · high-signal regimes | estimation error | Hurt |
Closest papers by meaning
- A Distribution Free Conditional Independence Test with Applications to Causal Discovery (JMLR 2022 · similarity 0.87)
- Conditional Dependence via Shannon Capacity: Axioms, Estimators and Applications (ICML 2016 · similarity 0.87)
- Extracting Rare Dependence Patterns via Adaptive Sample Reweighting (ICML 2025 · similarity 0.87)
- PAIR-CI: Calibrated Conditional Independence Testing for Causal Discovery with Incomplete Data (NeurIPS 2026 · similarity 0.85)
- Permutation-Free High-Order Interaction Tests (ICML 2025 · similarity 0.85)
- On the sample complexity of conditional independence testing with Von Mises estimator with application to causal discovery (ICML 2024 · similarity 0.85)
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.
