Research map / Reinforcement learning / Policy gradient and actor-critic
Lyapunov Density Models: Constraining Distribution Shift in Learning-Based Control
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 |
|---|---|---|---|
| constrain MPC with an LDM instead of using no constraint | MPC with learned dynamics · Hopper | average reward | Helped |
| constrain MPC with an LDM instead of using no constraint | MPC with learned dynamics · Lunar Lander | average reward | Helped |
| constrain MPC with an LDM instead of using no constraint | MPC with learned dynamics · SimGlucose | average reward | Helped |
| use a density-model constraint instead of no constraint | MPC with learned dynamics · Hopper, Lunar Lander, and SimGlucose | average reward | Helped |
| use an ensemble-variance constraint instead of no constraint | MPC with learned dynamics · Hopper, Lunar Lander, and SimGlucose | average reward | Helped |
| increase the LDM constraint threshold from low to medium | LDM-constrained MPC · Hopper | failure rate | Helped |
| increase the LDM constraint threshold from medium to high | LDM-constrained MPC · Hopper | average reward | Hurt |
| use a density-model constraint rather than an LDM constraint | MPC with learned dynamics · Hopper, Lunar Lander, and SimGlucose | average reward | Hurt |
| use an ensemble-variance constraint rather than an LDM constraint | MPC with learned dynamics · Hopper, Lunar Lander, and SimGlucose | average reward | Hurt |
Closest papers by meaning
- C-Learning: Learning to Achieve Goals via Recursive Classification (ICLR 2021 · similarity 0.85)
- Safe and Stable Control via Lyapunov-Guided Diffusion Models (NeurIPS 2025 · similarity 0.85)
- Analytical Lyapunov Function Discovery: An RL-based Generative Approach (ICML 2025 · similarity 0.84)
- Lyapunov-stable Neural Control for State and Output Feedback: A Novel Formulation (ICML 2024 · similarity 0.84)
- Safely Learning Controlled Stochastic Dynamics (NeurIPS 2025 · similarity 0.83)
- Learning to Stabilize Online Reinforcement Learning in Unbounded State Spaces (ICML 2024 · similarity 0.83)
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.