Research map / Bandits and online learning / Multi-armed bandits
Optimal Best-arm Identification in Linear Bandits
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 Lazy TS instead of RAGE | Finite-arm best-arm identification · Many arms, K=1000–7500 | sample complexity: 1206.55 vs 1148.45; 1253.60 vs 1440.75; 1247.05 vs 1540.3; 1296.55 vs 1598.0 | Mixed |
| Use Lazy TS instead of RAGE with the improved stopping threshold | Finite-arm best-arm identification · Many arms, K=1000–10000 | sample complexity: 424.5 vs 1148.45; 458.15 vs 1440.75; 434.65 vs 1540.3; 448.0 vs 1598.0; 452.85 vs 1479.4 | Helped |
| Use Lazy TS instead of the Oracle stopping procedure | Finite-arm best-arm identification · Many arms, K=1000–10000 | sample complexity: 424.5 vs 476.45; 458.15 vs 492.15; 434.65 vs 515.6; 448.0 vs 547.65; 452.85 vs 564.85 | Mixed |
| Remove averaging and track only the current allocation (ablation) | Lazy TS · Many arms, K=1000–10000 | sample complexity: 424.5 vs 424.5; 458.15 vs 455.95; 434.65 vs 433.6; 448.0 vs 447.45; 452.85 vs 452.95 | No clear effect |
| Remove averaging and track only the current allocation (ablation) | Lazy TS · Many arms, K=1000–7500 (main setup) | sample complexity: 1409 -> 1206.55; 1404 -> 1253.60; 1401 -> 1247.05; 1434 -> 1296.55 | Helped |
Closest papers by meaning
- Best Arm Identification in Generalized Linear Bandits via Hybrid Feedback (NeurIPS 2026 · similarity 0.94)
- Almost Minimax Optimal Best Arm Identification in Piecewise Stationary Linear Bandits (NeurIPS 2024 · similarity 0.94)
- Optimal Multi-Fidelity Best-Arm Identification (NeurIPS 2024 · similarity 0.94)
- Breaking the $\log(1/\Delta_2)$ Barrier: Better Batched Best Arm Identification with Adaptive Grids (ICLR 2025 · similarity 0.93)
- Balancing Performance and Costs in Best Arm Identification (NeurIPS 2025 · similarity 0.93)
- Near Optimal Best Arm Identification for Clustered Bandits (ICML 2025 · similarity 0.93)
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.