Research map / Reinforcement learning / Policy gradient and actor-critic
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 ENODE dynamics instead of modified PETS | Model-based RL on noisy Pendulum, CartPole, and Acrobot · Pendulum, CartPole, Acrobot; noisy data with varied observation spacing | reward | Mixed |
| Use modified PETS, which includes the time increment as an input, instead of vanilla PETS | PETS dynamics model · CartPole; irregularly sampled observations | one-step prediction error | Helped |
| Use ENODE rather than discrete-time models | Model-based RL on Pendulum, CartPole, and Acrobot · Pendulum, CartPole, Acrobot; noise-free regularly sampled observations | reward | No clear effect |
| Use Deep PILCO rather than the proposed model variants | Model-based RL · Pendulum, CartPole, Acrobot; noise-free regularly sampled observations | rounds to solve | Hurt |
| Use shorter or longer imagination horizons instead of an intermediate horizon (ablation) | ENODE · CartPole; noise-free data with fixed observation intervals | reward | Hurt |
| Vary the number of ensemble members (ablation) | ENODE · CartPole; noise-free data with fixed observation intervals | reward | No clear effect |
| Add observation noise | ENODE and modified PETS · Pendulum, CartPole, Acrobot | rounds to solve | Hurt |
| Reduce the mean observation time difference | Modified PETS · Pendulum, CartPole, Acrobot; noisy data | reward | Hurt |
| Use stochastic rather than fixed observation-time increments | ENODE and modified PETS · Pendulum, CartPole, Acrobot; noisy data | reward | No clear effect |
| Use continuous-time trajectory integration rather than discrete transitions | CartPole control simulation · CartPole | state-solution error | Helped |
Closest papers by meaning
- Robust Model Based Reinforcement Learning Using $\mathcal{L}_1$ Adaptive Control (ICLR 2024 · similarity 0.89)
- Overcoming Slow Decision Frequencies in Continuous Control: Model-Based Sequence Reinforcement Learning for Model-Free Control (ICLR 2025 · similarity 0.88)
- Improving planning and MBRL with temporally-extended actions (NeurIPS 2025 · similarity 0.88)
- Continuous Q-Score Matching: Diffusion Guided Reinforcement Learning for Continuous-Time Control (NeurIPS 2025 · similarity 0.88)
- When to Sense and Control? A Time-adaptive Approach for Continuous-Time RL (NeurIPS 2024 · similarity 0.88)
- Langevin Soft Actor-Critic: Efficient Exploration through Uncertainty-Driven Critic Learning (ICLR 2025 · 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.