Research map / Probabilistic and scientific modeling / Dynamical systems
Efficient Exploration in Continuous-time Model-based 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 continuous-time OCoRL with adaptive measurement selection rather than discrete-time zero-order-hold control | OCoRL · All tested environments | cost | Helped |
| Use continuous-time OCoRL with equidistant sampling rather than discrete-time zero-order-hold control | OCoRL · 7 of 9 systems | cost | Helped |
| Replace equidistant measurements with adaptive measurement selection (ablation) | OCoRL · All systems tested | cost | Helped |
| Plan optimistically over plausible dynamics instead of greedily using the learned mean model (ablation) | OCoRL · Pendulum, all MSSs | cost | Helped |
| Use Greedy Max Determinant measurement selection (ablation) | OCoRL · Furuta Pendulum | cost: $23.64 \pm 0.22$ -> $\mathbf{23.52 \pm 0.18}$ | Helped |
| Use Greedy Max Determinant measurement selection (ablation) | OCoRL · Cart Pole | cost: $\mathbf{17.52 \pm 0.04}$ -> $17.53 \pm 0.04$ | Hurt |
| Use Greedy Max Determinant measurement selection (ablation) | OCoRL · Cancer Treatment | cost: $20.70 \pm 0.06$ -> $\mathbf{20.68 \pm 0.05}$ | Helped |
| Use Greedy Max Determinant measurement selection (ablation) | OCoRL · Glucose in Blood | cost: $\mathbf{15.23 \pm 0.01}$ -> $15.24 \pm 0.01$ | Hurt |
Closest papers by meaning
- Sample-efficient and Scalable Exploration in Continuous-Time RL (ICLR 2026 · similarity 0.93)
- Efficient Model-Based Reinforcement Learning Through Optimistic Thompson Sampling (ICLR 2025 · similarity 0.90)
- Beyond Optimism: Exploration With Partially Observable Rewards (NeurIPS 2024 · similarity 0.89)
- Bayesian Exploration Networks (ICML 2024 · similarity 0.89)
- Efficient Exploration in Average-Reward Constrained Reinforcement Learning: Achieving Near-Optimal Regret With Posterior Sampling (ICML 2024 · similarity 0.89)
- NeoRL: Efficient Exploration for Nonepisodic RL (NeurIPS 2024 · 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.