Research map / Learning theory and algorithms / Prediction and decision losses
Adaptive Identification of Populations with Treatment Benefit in Clinical Trials: Machine Learning Challenges and Solutions
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 AdaGCPI (uniform sampling of active set, successive elimination) instead of GSDS fixed-subpopulation design | Adaptive enrichment trial simulation · Bernoulli trial simulation, 3 subgroups, Table 1 | % successful trials | Mixed |
| Discard subgroups with effects below theta_min via futility removal (rather than keeping GSDS-style power to detect small effects) | AdaGGI and AdaGCPI vs GSDS · Original I-SPY-style experiment theta approx [0,0.05,0.1] | % successful trials: GSDS 67% vs AdaGGI 13%, AdaGCPI 7% | Hurt |
| Use LCB sampling (E_LCB) instead of UCB in AdaGGI (ablation) | AdaGGI on stylized Gaussian simulations · Stylized simulations K=10 | time to first good group identification | Helped |
| Use UCB sampling in AdaGGI when good groups have very different effects (ablation) | AdaGGI on stylized simulations (Scen. 1 means 0.5 and 1) · Stylized simulations | time to identify good groups | Helped |
| Add population-level removal criterion R_pop-fut to AdaGCPI (ablation) | AdaGCPI on stylized simulations · Stylized simulations | time to termination and time to discard bad groups | Helped |
| Replace Bonferroni-corrected anytime CI with tighter intervals (not tested; noted as limitation) | AdaGGI and AdaGCPI · Simulated trials | % successful trials | Hurt |
Closest papers by meaning
- Dynamic Subgroup Identification in Covariate-adjusted Response-adaptive Randomization Experiments (NeurIPS 2024 · similarity 0.88)
- Using Surrogates in Covariate-adjusted Response-adaptive Randomization Experiments with Delayed Outcomes (NeurIPS 2024 · similarity 0.86)
- Boosting Test Performance with Importance Sampling--a Subpopulation Perspective (AAAI 2025 · similarity 0.86)
- Active Adaptive Experimental Design for Treatment Effect Estimation with Covariate Choice (ICML 2024 · similarity 0.86)
- Optimistic Algorithms for Adaptive Estimation of the Average Treatment Effect (ICML 2025 · similarity 0.86)
- Safe Exploration in Dose Finding Clinical Trials with Heterogeneous Participants (ICML 2024 · similarity 0.86)
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.