Research map / Bandits and online learning / RL theory and MDPs
Unifying PAC and Regret: Uniform PAC Bounds for Episodic Reinforcement Learning
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 finite-time law-of-iterated-logarithm confidence bounds instead of episode-dependent confidence widths | Optimism-based episodic RL · Randomly generated episodic MDPs | expected return | Helped |
| Use time-uniform confidence bounds and optimistic planning in UBEV | Finite-state episodic MDPs · Finite-state episodic MDPs | Uniform-PAC sample complexity | Helped |
| Use Uniform-PAC mistake bounds | Episodic RL algorithms · Finite-state episodic MDPs | high-probability regret | Helped |
Closest papers by meaning
- Hybrid Reinforcement Learning Breaks Sample Size Barriers In Linear MDPs (NeurIPS 2024 · similarity 0.87)
- Achieving Constant Regret in Linear Markov Decision Processes (NeurIPS 2024 · similarity 0.87)
- DARLING: Detection Augmented Reinforcement Learning with Non-Stationary Guarantees (NeurIPS 2026 · similarity 0.87)
- A PAC Learning Algorithm for LTL and Omega-Regular Objectives in MDPs (AAAI 2024 · similarity 0.87)
- No-Regret Reinforcement Learning in Smooth MDPs (ICML 2024 · similarity 0.87)
- Uniform Last-Iterate Guarantee for Bandits and Reinforcement Learning (NeurIPS 2024 · similarity 0.87)
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.