Research map / Probabilistic and scientific modeling / Treatment effects
Causal-BALD: Deep Bayesian Active Learning of Outcomes to Infer Treatment-Effects from Observational 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 μρBALD acquisition instead of random, propensity, or γ acquisition | DUE-based CATE estimation · Synthetic, IHDP, and CMNIST | sqrt PEHE | Helped |
| Add overlap-aware information gain to μBALD, forming μρBALD (ablation) | CATE active learning · Synthetic, IHDP, and CMNIST | sqrt PEHE | Helped |
| Use μπBALD, which combines μBALD with propensity-score weighting (ablation) | CATE active learning · High-dimensional covariates | sqrt PEHE | Hurt |
| Use ρBALD, which conditions information gain on the CATE (ablation) | CATE active learning · Increasing covariate dimensionality | sqrt PEHE | Helped |
| Use μBALD, based on reducible uncertainty in the observed outcome (ablation) | CATE active learning · Synthetic, IHDP, and CMNIST | sqrt PEHE | Helped |
| Use τBALD based on uncertainty in the treatment-effect estimate (ablation) | CATE active learning · Regions with limited or no overlap | sample efficiency | Hurt |
Closest papers by meaning
- Enhancing Treatment Effect Estimation via Active Learning: A Counterfactual Covering Perspective (ICML 2025 · similarity 0.88)
- Learning Shadow Variable Representation for Treatment Effect Estimation under Collider Bias (ICML 2024 · similarity 0.86)
- ABC3: Active Bayesian Causal Inference with Cohn Criteria in Randomized Experiments (AAAI 2025 · similarity 0.85)
- Improving the Generation and Evaluation of Synthetic Data for Downstream Medical Causal Inference (NeurIPS 2025 · similarity 0.85)
- Controllable Generative Sandbox for Causal Inference (NeurIPS 2026 · similarity 0.85)
- SEL-BALD: Deep Bayesian Active Learning with Selective Labels (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.