Research map / Bandits and online learning / RL theory and MDPs
Computationally Efficient Horizon-Free Reinforcement Learning for Linear Mixture MDPs
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 variance- and uncertainty-aware weights instead of variance-only weights | Algorithm bandit versus WeightedOFUL · Heterogeneous linear bandits | regret bound: \tilde O(d\sqrt{\sum_{k=1}^K\sigma_k^2} + \sqrt{dK} + d) -> \tilde O(d\sqrt{\sum_{k=1}^K\sigma_k^2} + d) | Helped |
| Use Algorithm bandit's estimator rather than VOFUL2 | Heterogeneous linear bandits · Heterogeneous linear bandits | regret bound: \tilde O(d^{1.5}\sqrt{\sum_{k=1}^K\sigma_k^2} + d^2) -> \tilde O(d\sqrt{\sum_{k=1}^K\sigma_k^2} + d) | Helped |
| Use Algorithm MDP rather than VARLin | Linear mixture MDPs · Linear mixture MDPs | regret bound: \tilde O(d^{4.5}\sqrt{K}+d^9) -> \tilde O(d\sqrt{K} + d^2) | Helped |
| Make the horizon-free VTR algorithm computationally efficient | Algorithm MDP versus VARLin2 · Linear mixture MDPs | regret bound: \tilde O(d\sqrt{K}+d^2) -> \tilde O(d\sqrt{K} + d^2) | No clear effect |
| Use high-order recursive moment regression | Algorithm MDP · Linear mixture MDPs | regret bound | Helped |
Closest papers by meaning
- Horizon-free Reinforcement Learning in Adversarial Linear Mixture MDPs (ICLR 2024 · similarity 0.92)
- Horizon-Free Regret for Linear Markov Decision Processes (ICLR 2024 · similarity 0.91)
- A Computationally Efficient Algorithm for Infinite-Horizon Average-Reward Linear MDPs (ICML 2025 · similarity 0.89)
- Offline-Online Reinforcement Learning for Linear Mixture MDPs (NeurIPS 2026 · similarity 0.89)
- Near-Optimal Dynamic Regret for Adversarial Linear Mixture MDPs (NeurIPS 2024 · similarity 0.88)
- A Primal-Dual Algorithm for Offline Constrained Reinforcement Learning with Linear MDPs (ICML 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.