Research map / Probabilistic and scientific modeling / Treatment effects
Efficient Online Estimation of Causal Effects by Deciding What to Observe
Shantanu Gupta, Zachary Chase Lipton, David Childers
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 OMS-ETG instead of OMS-ETC | OMS online moment selection · linear IV graph, T = 300 | relative regret | Improved |
| increase exploration fraction in ETC from 0.1/0.2 to larger | OMS-ETC · linear IV graph, small horizon | relative regret | Improved |
| use unregularized GMM weight matrix (lambda_W = 0) instead of 0.01 (ablation) | OMS-ETG · synthetic linear IV | regret: 1.98% | No clear effect |
| use OMS-ETG instead of OMS-ETC with 40% exploration | two IVs graph, kappa* at simplex corner · two IVs graph | relative regret | Improved |
| use adaptive policy instead of fixed_equal | OMS methods · confounder-mediator graph | relative regret: 20% | Improved |
| OMS-ETG-FB instead of OMS-ETG-FS | OMS-ETG · confounder-mediator graph | relative regret | No clear effect |
| use adaptive policy instead of collect_all | OMS methods · IHDP semi-synthetic | relative regret: 50% | Improved |
| increase ETC exploration from 0.2 to 0.4 | OMS-ETC · IHDP semi-synthetic | relative regret | Improved |
| use adaptive policy instead of fixed_equal | OMS methods · Vietnam draft earnings semi-synthetic | relative regret: 25% | Improved |
| use OMS-ETG instead of OMS-ETC | OMS · Vietnam draft earnings semi-synthetic, small horizons | relative regret | Improved |
| increase horizon T | all adaptive OMS policies · linear IV graph | relative regret | Improved |
Closest papers by meaning
- Local Learning for Covariate Selection in Nonparametric Causal Effect Estimation with Latent Variables (NeurIPS 2025 · similarity 0.88)
- Data Fusion for Partial Identification of Causal Effects (NeurIPS 2025 · similarity 0.88)
- Optimistic Algorithms for Adaptive Estimation of the Average Treatment Effect (ICML 2025 · similarity 0.88)
- Data-Driven Covariate Selection for Nonparametric and Cycle-Agnostic Causal Effect Estimation (NeurIPS 2026 · similarity 0.88)
- Empirical Analysis of Model Selection for Heterogeneous Causal Effect Estimation (ICLR 2024 · similarity 0.87)
- Off-policy estimation with adaptively collected data: the power of online learning (NeurIPS 2024 · 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.
