Research map / Federated learning and privacy / Differential privacy
Differentially Private Nonlinear Causal Discovery from Numerical Data
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 RCIT with KRR and HSIC permutation approximation instead of conditional Kendall’s τ | Private independence testing · Synthetic independence tests | Type II error rate | Helped |
| Use RCIT with KRR and HSIC permutation approximation instead of conditional Kendall’s τ | Private causal discovery with PCD · Cancer, Asia, Child, and Alarm simulations | Recall | Helped |
| Use RCIT with KRR and HSIC permutation approximation instead of conditional Kendall’s τ | Private causal discovery with PCD · Sachs | F1 | Helped |
| Increase the privacy parameter ϵ | RCIT · Synthetic conditional independence tests | Type I error rate | Helped |
| Increase sample size from 200 to 500 | RCIT · Synthetic conditional independence tests | Type I error rate | Helped |
| Apply conditional Kendall’s τ to the tested cases | Private conditional independence testing · Synthetic conditional independence tests | Type I error rate | Hurt |
| Use RCIT instead of conditional Kendall’s τ | Private independence testing · Synthetic independence tests with 500 samples | Type I error rate | Hurt |
| Use private RCIT instead of non-private RCIT | RCIT · Synthetic conditional independence tests | Type I error rate | Hurt |
| Use conditional Kendall’s τ instead of RCIT | Private causal discovery with PCD · Sachs | Recall | Hurt |
Closest papers by meaning
- Causal Discovery via Conditional Independence Testing with Proxy Variables (ICML 2024 · similarity 0.88)
- On the sample complexity of conditional independence testing with Von Mises estimator with application to causal discovery (ICML 2024 · similarity 0.88)
- Federated Causal Discovery from Heterogeneous Data (ICLR 2024 · similarity 0.87)
- Differentiable Constraint-Based Causal Discovery (NeurIPS 2025 · similarity 0.87)
- Causal Discovery with Fewer Conditional Independence Tests (ICML 2024 · similarity 0.87)
- Signature Kernel Conditional Independence Tests in Causal Discovery for Stochastic Processes (ICLR 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.