Research map / Probabilistic and scientific modeling / Treatment effects
Causal-EPIG: Causally Aligned Active CATE Estimation
Erdun Gao, Jake Fawkes, Dino Sejdinovic
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 |
|---|---|---|---|
| target the CATE directly (Causal-EPIG-tau) instead of factual/parameter proxies | active CATE estimation · Synthetic datasets (180 dataset-seed combos) | average rank: 3.03 vs Random 6.52 | Improved |
| target the joint potential-outcome pair (Causal-EPIG-mu) instead of proxies | active CATE estimation · Synthetic datasets (180 dataset-seed combos) | average rank: 3.98 vs Random 6.52 | Improved |
| use Causal-EPIG-tau instead of Causal-EPIG-mu | active CATE estimation · Synthetic datasets (180 dataset-seed combos) | average rank: 3.98 -> 3.03 | Improved |
| use factual-outcome EPIG instead of causal-aligned targets | EPIG · Synthetic datasets (180 dataset-seed combos) | average rank: 6.92 vs 3.03 | Hurt |
| use Causal-EPIG-mu with BCF estimator | Causal-EPIG-mu · Synthetic (BCF) | sqrt PEHE | Mixed |
| use direct CATE targeting when treatment effect cancels in the arm difference (cancellation setting) (ablation) | Causal-EPIG-tau · Targeted synthetic (cancellation) | sqrt PEHE | Improved |
| use full PO targeting when effect arises from arm-surface differences (surface-split) (ablation) | Causal-EPIG-mu · Targeted synthetic (surface-split) | sqrt PEHE | Improved |
| use propensity-based acquisition on IHDP | Propensity baseline · IHDP | sqrt PEHE | No clear effect |
Closest papers by meaning
- Optimistic Algorithms for Adaptive Estimation of the Average Treatment Effect (ICML 2025 · similarity 0.89)
- Empirical Analysis of Model Selection for Heterogeneous Causal Effect Estimation (ICLR 2024 · similarity 0.88)
- Do Contemporary Causal Inference Models Capture Real-World Heterogeneity? Findings from a Large-Scale Benchmark (ICLR 2025 · similarity 0.88)
- ABC3: Active Bayesian Causal Inference with Cohn Criteria in Randomized Experiments (AAAI 2025 · similarity 0.88)
- CausalPFN: Amortized Causal Effect Estimation via In-Context Learning (NeurIPS 2025 · similarity 0.88)
- Treatment Effect Estimation for Optimal Decision-Making (NeurIPS 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.
