Research map / Probabilistic and scientific modeling / Treatment effects
Stable Estimation of Heterogeneous Treatment Effects
Anpeng Wu, Kun Kuang, Ruoxuan Xiong, Bo Li, Fei Wu
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 |
|---|---|---|---|
| Uniform resampling + epsilon-greedy matching (StableCFR) vs. no imbalance technique (VANILLA) | two-head network on Syn-0.8 · Syn-0.8 OOD data | PEHE: 0.500 -> 0.299 | Improved |
| Uniform resampling + epsilon-greedy matching (StableCFR) | two-head network on Syn-0.8 · Syn-0.8 ID data | PEHE: 0.089 -> 0.099 | Hurt |
| Uniform resampling + epsilon-greedy matching (StableCFR) | two-head network on Semi-PM-CMR · Semi-PM-CMR OOD data | PEHE: 0.548 (IPWNet) -> 0.489; about 11% reduction vs SOTA | Improved |
| Uniform resampling + epsilon-greedy matching (StableCFR) | two-head network on Semi-PM-CMR · Semi-PM-CMR ID data | PEHE: 0.131 (VANILLA) vs 0.139 | No clear effect |
| Uniform resampling + epsilon-greedy matching (StableCFR) | two-head network on Syn-0.8 · Syn-0.8 OOD data | MAXE: 7.628 (DeR-CFR) -> 5.889 | Improved |
| Pure uniform sampling with nearest neighbor (epsilon=1) vs. with epsilon-greedy matching (ablation) | StableCFR · Syn-0.8 (ablation of matching) | qualitative PEHE and variance | Improved |
| Conventional CFR-based balancing methods (CFRNet, SITE, CFR-ISW) vs. StableCFR without regularity | representation balancing baselines · Syn-0.8 OOD data | PEHE: CFRNet 5.990, SITE 7.008 vs StableCFR 0.299 | Hurt |
| Imbalance regression reweighting/balanced MSE (Reweight, BMC, GAI, DIRNet) vs. VANILLA | regression baselines on Syn-0.8 · Syn-0.8 OOD data | PEHE: Reweight 1.205, BMC 1.574, GAI 2.136, DIRNet 0.860 vs VANILLA 0.500 | Hurt |
Closest papers by meaning
- Heterogeneous Treatment Effect in Time-to-Event Outcomes: Harnessing Censored Data with Recursively Imputed Trees (ICML 2025 · similarity 0.86)
- A Non-parametric Direct Learning Approach to Heterogeneous Treatment Effect Estimation under Unmeasured Confounding (NeurIPS 2024 · similarity 0.86)
- Doubly robust identification of treatment effects from multiple environments (ICLR 2025 · similarity 0.85)
- Continuous Treatment Effect Estimation Using Gradient Interpolation and Kernel Smoothing (AAAI 2024 · similarity 0.85)
- Counterfactual Contrastive Learning with Normalizing Flows for Robust Treatment Effect Estimation (ICML 2025 · similarity 0.85)
- Hierarchical and Density-based Causal Clustering (NeurIPS 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.
