Research map / Bandits and online learning / Multi-armed bandits
Optimal Estimation of the Best Mean in Multi-Armed Bandits
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Studies estimating the mean reward of the best arm in a multi-armed bandit to within ε with confidence 1-δ using as few samples as possible, with an instance-dependent lower bound.
Problem
Estimating the largest mean reward among arms in a multi-armed bandit to ε accuracy with probability at least 1-δ while minimizing sample count.
Approach
Establishes an instance-dependent sample complexity lower bound for estimating the best arm's mean; the abstract states it is followed by optimal estimation (details not available in the provided text).
Experimental setup
Metrics: sample complexity
Closest papers by meaning
- A Closer Look at the Worst-case Behavior of Multi-armed Bandit Algorithms (NeurIPS 2021 · similarity 0.92)
- On the Asymptotic Optimality of Confidence Interval Based Algorithms for Fixed Confidence MABs (AAAI 2025 · similarity 0.91)
- Best Arm Identification in Generalized Linear Bandits via Hybrid Feedback (NeurIPS 2026 · similarity 0.90)
- Fixed Confidence Best Arm Identification in the Bayesian Setting (NeurIPS 2024 · similarity 0.90)
- Balancing Performance and Costs in Best Arm Identification (NeurIPS 2025 · similarity 0.90)
- Forced Exploration in Bandit Problems (AAAI 2024 · similarity 0.90)
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.