Research map / Bandits and online learning / RL theory and MDPs
Learning Unknown Markov Decision Processes: A Thompson Sampling Approach
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 |
|---|---|---|---|
| Add the episode-length growth stopping criterion to the doubling-based schedule (ablation) | TSDE · RiverSwim | regret | Helped |
| Use TSDE rather than UCRL2, TSMDP, or Lazy PSRL | Infinite-horizon MDP learning · Randomly generated MDPs | regret | Helped |
| Use TSDE rather than UCRL2, Lazy PSRL, or TSMDP with s₀=3 | Infinite-horizon MDP learning · RiverSwim | regret | Helped |
| Use TSMDP with s₀=1 rather than TSDE | RiverSwim · RiverSwim | regret | Helped |
| Use TSMDP with a fixed resampling state in a randomly generated MDP | TSMDP · Randomly generated MDPs | regret | Hurt |
| Use ε_k-approximate policies with ε_k ≤ 1/(k+1) instead of exact episode policies | TSDE · Weakly communicating MDPs, Bayesian regret through time T | expected regret: \tilde O(HS\sqrt{AT}) | No clear effect |
Closest papers by meaning
- Optimistic posterior sampling for reinforcement learning: worst-case regret bounds (NeurIPS 2017 · similarity 0.92)
- No-Regret Thompson Sampling for Finite-Horizon Markov Decision Processes with Gaussian Processes (NeurIPS 2025 · similarity 0.92)
- Efficient Exploration in Average-Reward Constrained Reinforcement Learning: Achieving Near-Optimal Regret With Posterior Sampling (ICML 2024 · similarity 0.90)
- Bayesian Learning of Optimal Policies in Markov Decision Processes with Countably Infinite State-Space (NeurIPS 2023 · similarity 0.90)
- Sample Efficient Reinforcement Learning with Partial Dynamics Knowledge (AAAI 2024 · similarity 0.89)
- DARLING: Detection Augmented Reinforcement Learning with Non-Stationary Guarantees (NeurIPS 2026 · similarity 0.89)
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.