Research map / Probabilistic and scientific modeling / Treatment effects
Sinkhorn Treatment Effects: A Causal Optimal Transport Measure
Medha Agarwal, Alex Luedtke
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 Sinkhorn treatment effect test instead of MMD-based MTE test | second-order one-step test · PCam simulation (n = 5000) | type I error at alpha 0.05: 0.11 -> 0.04 | Improved |
| use STE instead of MTE test | second-order one-step test · PCam simulation | power | Hurt |
| add one-step debiasing to the plug-in STE estimator (ablation) | STE estimator · Mean shift simulation (theta = 1.6) | mean squared error | Improved |
| increase sample size | STE and MTE tests · Gaussian simulation | type I error and power | Improved |
| use STE instead of MTE for confidence intervals | Wald 95% CI · Mean shift simulation (theta = 1.6) | coverage | No clear effect |
Closest papers by meaning
- Conditional Counterfactual Mean Embeddings: Doubly Robust Estimation and Learning Rates (NeurIPS 2026 · similarity 0.86)
- Reducing Balancing Error for Causal Inference via Optimal Transport (ICML 2024 · similarity 0.85)
- Semiparametric Efficient Tests for Interpretable Distributional Treatment Effects (NeurIPS 2026 · similarity 0.85)
- Generalization through variance: how noise shapes inductive biases in diffusion models (ICLR 2025 · similarity 0.85)
- Doubly-Robust Estimation of Counterfactual Policy Mean Embeddings (NeurIPS 2025 · similarity 0.84)
- Reducing Confounding Bias without Data Splitting for Causal Inference via Optimal Transport (ICML 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.
