Research map / Bandits and online learning / Multi-armed bandits
Dealing With Misspecification In Fixed-Confidence Linear Top-m Identification
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 LinGapE on simulated instances with large misspecification (ε=5) | LinGapE · Simulated misspecified instances | error frequency: 0.96 | Hurt |
| Use LinGapE with its theoretically supported stopping threshold | LinGapE · Simulated misspecified instances (ε=5) | error frequency: 1 | Hurt |
| Use LUCB instead of structured methods | LUCB · Simulated misspecified instances | sample complexity | Hurt |
| Increase the user-provided misspecification bound ε | MisLid · Discrepancy between user-selected ε and true ‖η‖ | sample complexity | Hurt |
| Set ε large enough to reach the unstructured regime | MisLid · Discrepancy between user-selected ε and true ‖η‖ | sample complexity | Mixed |
| Remove optimism and use empirical gains (ablation) | MisLid · Comparing different optimisms | sample complexity | Helped |
| Use empirical gains instead of optimism (ablation) | MisLid · Online recommendation (Last.fm) | sample complexity | Helped |
| Use MisLid rather than an unstructured bandit algorithm | Drug repurposing in epilepsy · Drug repurposing in epilepsy | sample complexity: an order of magnitude | Helped |
| Use MisLid or LinGapE rather than LUCB | Drug repurposing in epilepsy · Drug repurposing in epilepsy | sample complexity | Helped |
| Remove the optimistic bonus from sampling gains (ablation) | MisLid · General experimental evaluation | sample complexity | Helped |
Closest papers by meaning
- Fixed Confidence Best Arm Identification in the Bayesian Setting (NeurIPS 2024 · similarity 0.87)
- Best Arm Identification in Generalized Linear Bandits via Hybrid Feedback (NeurIPS 2026 · similarity 0.87)
- Balancing Performance and Costs in Best Arm Identification (NeurIPS 2025 · similarity 0.86)
- Optimal Multi-Fidelity Best-Arm Identification (NeurIPS 2024 · similarity 0.86)
- On the Asymptotic Optimality of Confidence Interval Based Algorithms for Fixed Confidence MABs (AAAI 2025 · similarity 0.85)
- Misspecified $Q$-Learning with Sparse Linear Function Approximation: Tight Bounds on Approximation Error (ICLR 2025 · similarity 0.85)
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.