Research map / Bandits and online learning / Multi-armed bandits
Batched Multi-armed Bandits Problem
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 BaSE with a minimax static grid | Batched K-armed bandit · Minimax regret | expected cumulative regret: C\log K\sqrt{\log(KT)}\cdot \sqrt{K}T^{\frac{1}{2-2^{1-M}}} | Helped |
| Use BaSE with a geometric static grid | Batched K-armed bandit · Problem-dependent regret | expected cumulative regret: C\log K\log (KT)\cdot \frac{KT^{1/M}}{\min_{i\neq \star} \Delta_i} | Helped |
| Use an adaptive grid rather than a static grid | Batched K-armed bandit · Minimax and problem-dependent regret | regret lower bound: cM^{-2}\cdot \sqrt{K}T^{\frac{1}{2-2^{1-M}}}; cM^{-2}\cdot KT^{\frac{1}{M}} | No clear effect |
| Use the minimax grid instead of the geometric or arithmetic grid | BaSE · Experimental bandit settings | average regret | Helped |
| Increase the number of batches to 4 | BaSE · Experimental bandit settings | average regret relative to centralized UCB1 | Helped |
| Use BaSE instead of ETC | Two-arm batched bandit · Experiment comparing BaSE and ETC | regret | Helped |
Closest papers by meaning
- Regret Bounds for Batched Bandits (AAAI 2021 · similarity 0.94)
- Optimal and Practical Batched Linear Bandit Algorithm (ICML 2025 · similarity 0.91)
- Optimal Batched Linear Bandits (ICML 2024 · similarity 0.89)
- The Batch Complexity of Bandit Pure Exploration (ICML 2025 · similarity 0.89)
- Batched Stochastic Linear Bandits with 1-Bit Communication Constraints (NeurIPS 2026 · similarity 0.89)
- Breaking the $\log(1/\Delta_2)$ Barrier: Better Batched Best Arm Identification with Adaptive Grids (ICLR 2025 · 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.