Research map / Reinforcement learning / MDPs and planning
The Regret of Exploration and the Control of Bad Episodes in 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 |
|---|---|---|---|
| add performance test (PT) to end episodes when current policy's optimistic value falls below optimistic optimum by sqrt(alpha log t / t) | UCRL2-PT · DMDP non-degenerate, Bernoulli rewards | regret of exploration | Helped |
| add performance test (PT) to end episodes | UCRL2, UCRL2B, KL-UCRL · RiverSwim and random MDP | regret | No clear effect |
| use sparse PT testing and memorisation to accelerate PT variants (ablation) | UCRL2 with PT · 5-state RiverSwim | iterations per second: from 1.0k (pure PT) to 82k (PT+M+S) | Helped |
| set alpha large (slow decay of bonus threshold) (ablation) | UCRL2-PT · RiverSwim | frequency of PT triggering | Hurt |
| set alpha to zero (always optimistic policy, UCB-like) (ablation) | UCRL2-PT · theory | regret | Hurt |
Closest papers by meaning
- Beyond Optimism: Exploration With Partially Observable Rewards (NeurIPS 2024 · similarity 0.90)
- Tighter Problem-Dependent Regret Bounds in Reinforcement Learning without Domain Knowledge using Value Function Bounds (ICML 2019 · similarity 0.90)
- Minimax Optimal Reinforcement Learning with Quasi-Optimism (ICLR 2025 · similarity 0.89)
- Position: Constants are Critical in Regret Bounds for Reinforcement Learning (ICML 2025 · similarity 0.89)
- NeoRL: Efficient Exploration for Nonepisodic RL (NeurIPS 2024 · similarity 0.89)
- Minimax Optimal Regret Bound for Reinforcement Learning with Trajectory Feedback (ICML 2025 · 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.