Research map / Probabilistic and scientific modeling / Treatment effects
A Kernel Test for Causal Association via Noise Contrastive Backdoor Adjustment
Robert Hu, Dino Sejdinovic, Robin J. Evans
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 a linear kernel instead of RBF for univariate continuous treatment (ablation) | bd-HSIC · Simulation, continuous treatment | type 1 error calibration | Hurt |
| use a linear kernel for univariate binary treatment (ablation) | bd-HSIC · Simulation, binary treatment | power | Improved |
| replace estimated density-ratio weights with random uniform weights (ablation) | bd-HSIC · Simulation | power | Hurt |
| increase confounding strength beta_XZ so density ratio estimation fails (ablation) | bd-HSIC with NCE-p* and TRE-p* · Simulation, d_X=d_Y=3, d_Z=15, null | type 1 error | Hurt |
| compare bd-HSIC with PDS under symmetric non-linear X-Y dependency | PDS · Simulation, non-linear dependency | power | Hurt |
| use bd-CME with high-dimensional confounders | bd-CME · Simulation, d_Z=15 continuous and d_Z=50 mixed | type 1 error | Hurt |
Closest papers by meaning
- A Kernel-based Test of Independence for Cluster-correlated Data (NeurIPS 2021 · similarity 0.90)
- Independence Test for Linear Non-Gaussian Data and Applications in Causal Discovery (ICLR 2026 · similarity 0.87)
- Statistical Insights into HSIC in High Dimensions (NeurIPS 2023 · similarity 0.87)
- PAIR-CI: Calibrated Conditional Independence Testing for Causal Discovery with Incomplete Data (NeurIPS 2026 · similarity 0.86)
- Causal Discovery via Conditional Independence Testing with Proxy Variables (ICML 2024 · similarity 0.85)
- On the Hardness of Conditional Independence Testing In Practice (NeurIPS 2025 · 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.
