Research map / Bandits and online learning / Multi-armed bandits
An active learning framework for multi-group mean estimation
Abdellah Aznag, Rachel Cummings, Adam N. Elmachtoub
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 |
|---|---|---|---|
| vary p-norm over 1, 2, 10, 25, +inf (ablation) | Variance-UCB · Gaussian 2-group simulation | regret convergence slope: -2 for finite p, -1.5 for p = +inf | Mixed |
| overestimate C_T (1000 instead of 5) (ablation) | Variance-UCB · Gaussian 2-group simulation | regret | Hurt |
| underestimate C_T (0.001 instead of 5) (ablation) | Variance-UCB · Gaussian 2-group simulation | regret | Mixed |
| increase lowest standard deviation sigma_min from 0.05 to 1 (ablation) | Variance-UCB · finite p (1, 2) | regret | Improved |
| increase lowest variance sigma_min (ablation) | Variance-UCB · p = +inf | regret | No clear effect |
| increase number of groups G from 2 to 10 and 50 (ablation) | Variance-UCB · Gaussian groups | regret: ~30, 000 for G = 10 and ~90, 000 for G = 50 | Hurt |
Closest papers by meaning
- Robust Sampling for Active Statistical Inference (NeurIPS 2025 · similarity 0.84)
- Only Pay for What Is Uncertain: Variance-Adaptive Thompson Sampling (ICLR 2024 · similarity 0.84)
- Active Context Selection Improves Simple Regret in Contextual Bandits (NeurIPS 2026 · similarity 0.84)
- Adaptive Data Collection for Robust Learning Across Multiple Distributions (ICML 2025 · similarity 0.83)
- Group-wise oracle-efficient algorithms for online multi-group learning (NeurIPS 2024 · similarity 0.83)
- Generalized Linear Bandits: Almost Optimal Regret with One-Pass Update (NeurIPS 2025 · similarity 0.83)
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.
