Research map / Bandits and online learning / Multi-armed bandits
Multi-Fidelity Best-Arm Identification
Riccardo Poiani, Alberto Maria Metelli, Marcello Restelli
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 IISE instead of Successive Elimination | best-arm identification · Synthetic A | cost complexity (% of SE): 100 -> 11.47 | Improved |
| use IISE instead of Successive Elimination | best-arm identification · Synthetic B | cost complexity (% of SE): 100 -> 27.06 | Improved |
| use IISE instead of Successive Elimination | best-arm identification · Yahtzee | cost complexity (% of SE): 100 -> 0.53 | Improved |
| use order-aware bias variation gamma instead of max bias xi (IISE-γ) (ablation) | IISE · Synthetic A | cost complexity (% of SE): 11.47 -> 2.53 | Improved |
| use order-aware bias variation gamma instead of max bias xi (IISE-γ) (ablation) | IISE · Synthetic B | cost complexity (% of SE): 27.06 -> 6.39 | Improved |
| use order-aware bias variation gamma instead of max bias xi (IISE-γ) (ablation) | IISE · Yahtzee | cost complexity (% of SE): 0.53 -> 0.001 | Improved |
| choose thresholds alpha_m near 0 to exploit every fidelity (MFE) instead of IISE thresholds (ablation) | IISE · Synthetic A | cost complexity (% of SE): 11.47 -> 212.43 | Hurt |
| choose thresholds alpha_m near 0 to exploit every fidelity (MFE) instead of IISE thresholds (ablation) | IISE · Synthetic B | cost complexity (% of SE): 27.06 -> 3330.27 | Hurt |
| choose thresholds alpha_m near 0 to exploit every fidelity (MFE) instead of IISE thresholds (ablation) | IISE · Yahtzee | cost complexity (% of SE): 0.53 -> 342.75 | Hurt |
Closest papers by meaning
- Optimal Multi-Fidelity Best-Arm Identification (NeurIPS 2024 · similarity 0.91)
- Balancing Performance and Costs in Best Arm Identification (NeurIPS 2025 · similarity 0.88)
- Optimal Batched Best Arm Identification (NeurIPS 2024 · similarity 0.87)
- Nearly Optimal Fixed-Confidence Best-Arm Identification with 1-Bit Feedback (NeurIPS 2026 · similarity 0.87)
- Fixing the Loose Brake: Exponential-Tailed Stopping Time in Best Arm Identification (ICML 2025 · similarity 0.87)
- Breaking the $\log(1/\Delta_2)$ Barrier: Better Batched Best Arm Identification with Adaptive Grids (ICLR 2025 · similarity 0.86)
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.
