Research map / Bandits and online learning / Multi-armed bandits
Gamification of Pure Exploration for 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 W instead of DKM sampling | Linear BAI · Hard instance (d=2, α=0.1, δ=0.01) | average total pulls: 8705 -> 7135 | Helped |
| Use W-Cvx instead of DKM sampling | Linear BAI · Hard instance (d=2, α=0.1, δ=0.01) | average total pulls: 8705 -> 6854 | Helped |
| Use W-Cvx instead of W sampling | Linear BAI · Random unit-sphere instances (d=6, 8, 10, 12; δ=0.01) | empirical stopping time | Helped |
| Use W or W-Cvx rather than DKM sampling | Linear BAI · Hard instance (d=2, α=0.1, δ=0.01), arm a3 | average pulls: 1775 -> 104 (W); 1775 -> 77 (W-Cvx) | Helped |
| Target characteristic-time optimal weights rather than XY-optimal weights | Optimal-design lower-bound allocation · Hard instance (ΔMin= 0.0049958) | lower bound T_w: 2882 -> 369 | Helped |
| Target characteristic-time optimal weights rather than G-optimal weights | Optimal-design lower-bound allocation · Hard instance (ΔMin= 0.0049958) | lower bound T_w: 2882 -> 369 | Helped |
| Use W or W-Cvx rather than LGapE sampling | Linear BAI · Hard counterexample (d=2, α=π/6; δ=0.1, 0.01, 0.0001) | empirical stopping time | Hurt |
| Use W or W-Cvx rather than LGapE sampling | Linear BAI · Random unit-sphere instances (d=6, 8, 10, 12; δ=0.01) | empirical stopping time | Hurt |
Closest papers by meaning
- Instance-Optimal Pure Exploration for Linear Bandits on Continuous Arms (ICML 2025 · similarity 0.91)
- Best Arm Identification in Generalized Linear Bandits via Hybrid Feedback (NeurIPS 2026 · similarity 0.90)
- The Batch Complexity of Bandit Pure Exploration (ICML 2025 · similarity 0.90)
- Forced Exploration in Bandit Problems (AAAI 2024 · similarity 0.88)
- Near Optimal Non-asymptotic Sample Complexity of 1-Identification (ICML 2025 · similarity 0.88)
- Almost Minimax Optimal Best Arm Identification in Piecewise Stationary Linear Bandits (NeurIPS 2024 · similarity 0.88)
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.