Research map / Probabilistic and scientific modeling / Treatment effects
Should I Stop or Should I Go: Early Stopping with Heterogeneous Populations
Hammaad Adam, Fan Yin, Mary Hu, Neil Tenenholtz, Lorin Crawford, Lester Mackey, Allison Koenecke
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 |
|---|---|---|---|
| replace homogeneous stopping test with CLASH (heterogeneous weighting) | CLASH · Real A/B test | participants at stopping: 60,000 -> 40,000 | Improved |
| replace SUBTLE with CLASH | CLASH · Real A/B test | participants at stopping: 80,000 -> 40,000 | Improved |
| CLASH versus Oracle that only sees harmed region | CLASH · Real A/B test | participants at stopping: 40,000 -> 40,000 | No clear effect |
| apply CLASH to OF-adjusted Cox regression | CLASH · TTE simulation, 25% harmed, HR 0.7 | stopping probability: +20 percentage points | Improved |
| use CLASH instead of homogeneous baseline when minority harmed | CLASH · Gaussian simulation, 12.5% minority | stopping probability | Improved |
| use CLASH instead of SUBTLE at medium effect sizes | CLASH · Gaussian simulation | stopping probability | Improved |
| use CLASH when treatment has no harmful effect | CLASH · Gaussian simulation | Type I error / stopping probability | No clear effect |
| increase number of covariates (ablation) | CLASH · Gaussian simulation | stopping probability | No clear effect |
| vary hyperparameter delta (ablation) | CLASH · Gaussian simulation | stopping probability | No clear effect |
| increase minority group size (ablation) | CLASH · Gaussian simulation | stopping probability | Improved |
| very small sample (N=200) (ablation) | CLASH · Gaussian simulation | stopping probability | Hurt |
| very large covariate count (500) (ablation) | CLASH · Gaussian simulation | stopping probability | Hurt |
Closest papers by meaning
- Higher-Order Causal Message Passing for Experimentation with Complex Interference (NeurIPS 2024 · similarity 0.86)
- Hierarchical and Density-based Causal Clustering (NeurIPS 2024 · similarity 0.85)
- Do Contemporary Causal Inference Models Capture Real-World Heterogeneity? Findings from a Large-Scale Benchmark (ICLR 2025 · similarity 0.85)
- AutoCATE: End-to-End, Automated Treatment Effect Estimation (ICML 2025 · similarity 0.85)
- Data Fusion for Partial Identification of Causal Effects (NeurIPS 2025 · similarity 0.85)
- On Efficient Estimation of Distributional Treatment Effects under Covariate-Adaptive Randomization (ICML 2025 · 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.
