Research map / Bandits and online learning / RL theory and MDPs
Q-learning with UCB Exploration is Sample Efficient for Infinite-Horizon MDP
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 a UCB exploration bonus to Q-learning | Q-learning for infinite-horizon discounted MDPs without a generative model · Infinite-horizon discounted MDP (no generative model) | sample complexity of exploration: \tilde{O}({\frac{SA}{\epsilon^4(1-\gamma)^8}}) -> \tilde{O}({\frac{SA}{\epsilon^2(1-\gamma)^7}}) | Helped |
| Use Delayed Q-learning instead of UCB Q-learning | Q-learning for infinite-horizon discounted MDPs · Hard MDP family with constant S and A | sample complexity of exploration: \Omega\left(\frac{\epsilon^{-3}}{\ln(1/\delta)}\right) | Hurt |
Closest papers by meaning
- Tightening the Dependence on Horizon in the Sample Complexity of Q-Learning (ICML 2021 · similarity 0.92)
- Is Q-Learning Provably Efficient? (NeurIPS 2018 · similarity 0.92)
- Provably Efficient Q-Learning with Low Switching Cost (NeurIPS 2019 · similarity 0.90)
- UCB Momentum Q-learning: Correcting the bias without forgetting (ICML 2021 · similarity 0.90)
- Sample Efficient Reinforcement Learning with Partial Dynamics Knowledge (AAAI 2024 · similarity 0.89)
- A Nearly Optimal and Low-Switching Algorithm for Reinforcement Learning with General Function Approximation (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.