Research map / Bandits and online learning / Multi-armed bandits
An $\varepsilon$-Best-Arm Identification Algorithm for Fixed-Confidence and Beyond
Marc Jourdan, Rémy Degenne, Emilie Kaufmann
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 |
|---|---|---|---|
| Use EB-TCε with IDS and slack ε₀ = ε | ε-BAI on the α = 0.3 scenario · Gaussian bandits, up to K = 1000 | empirical stopping time | Improved |
| Use EB-TCε with fixed β = 1/2 and ε₀ = 0.1 | Anytime simple regret minimization · μ = (0.6,0.6,0.55,0.45,0.3,0.2) | simple regret | Improved |
| Use IDS proportions instead of fixed proportions β = 1/2 (ablation) | EB-TCε · Empirical stopping time and simple regret experiments | performance | Improved |
| Set the slack ε₀ equal to the target error ε rather than below it (ablation) | EB-TCε₀ · Fixed-confidence ε-BAI | empirical stopping time | Improved |
| Set the slack ε₀ below the target error ε (ablation) | EB-TCε₀ · Fixed-confidence ε-BAI | empirical stopping time | Hurt |
| Set the slack ε₀ too small (ablation) | EB-TCε₀ · Fixed-confidence identification | empirical performance | Hurt |
| Set the slack ε₀ too large (ablation) | EB-TCε₀ · Fixed-confidence identification | empirical performance | Hurt |
| Use a time-varying slack sequence εₙ decreasing to zero (ablation) | EB-TC with IDS · Best-arm identification (ε = 0) | empirical performance | Mixed |
| Add implicit exploration mechanisms such as a TS/UCB leader or TCI challenger (ablation) | EB-TCε · ε-BAI stopping-time experiment | empirical stopping time | No clear effect |
Closest papers by meaning
- Optimal Multi-Fidelity Best-Arm Identification (NeurIPS 2024 · similarity 0.92)
- Balancing Performance and Costs in Best Arm Identification (NeurIPS 2025 · similarity 0.92)
- Best Arm Identification in Generalized Linear Bandits via Hybrid Feedback (NeurIPS 2026 · similarity 0.91)
- Fixed-Budget Differentially Private Best Arm Identification (ICLR 2024 · similarity 0.91)
- Fixed Confidence Best Arm Identification in the Bayesian Setting (NeurIPS 2024 · similarity 0.91)
- Best Arm Identification for Stochastic Rising Bandits (ICML 2024 · similarity 0.91)
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.
