Research map / Bandits and online learning / RL theory and MDPs
Adaptive Sampling for Best Policy Identification in Markov Decision Processes
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 KLB-TS instead of BESPOKE | Best-policy identification · Random MDPs: 2 states × 2 actions and 5 states × 10 actions per state | sample complexity | Helped |
| Track the problem-specific allocation with KLB-TS | KLB-TS · Random MDPs, moderate confidence levels | sample complexity relative to 4 U(φ) log(1/δ): δ = 10^{-14} | Mixed |
| Use BESPOKE's adaptive allocation after its initialization | BESPOKE · Random MDPs | share of sample complexity attributable to initialization: more than 99% | Hurt |
Closest papers by meaning
- Optimal Sample Complexity for Average Reward Markov Decision Processes (ICLR 2024 · similarity 0.92)
- Near-Optimal Sample Complexity for MDPs via Anchoring (ICML 2025 · similarity 0.92)
- Finding good policies in average-reward Markov Decision Processes without prior knowledge (NeurIPS 2024 · similarity 0.92)
- Non-Asymptotic Best Policy Identification Guarantees in Online Reinforcement Learning (NeurIPS 2026 · similarity 0.90)
- Improved Lower Bounds for First-order Stochastic Non-convex Optimization under Markov Sampling (ICML 2025 · similarity 0.89)
- Span-Based Optimal Sample Complexity for Weakly Communicating and General Average Reward MDPs (NeurIPS 2024 · 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.