Research map / Reinforcement learning / Policy gradient and actor-critic
Constrained Update Projection Approach to Safe Policy Optimization
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 |
|---|---|---|---|
| CUP instead of FOCOPS | safe RL policy optimization · Hopper-v3 | average return: 1687.72 -> 2025.56 | Helped |
| CUP instead of FOCOPS | safe RL policy optimization · Ant-v3 | average return: 1662.53 -> 1743.66 | Helped |
| CUP instead of PPO-L | safe RL policy optimization · Swimmer-v3 | average return: 35.58 -> 33.38 | Hurt |
| CUP instead of PCPO | safe RL policy optimization · Swimmer-v3 | average return: 37.73 -> 33.38 | Hurt |
| vary step-size alpha (ablation) | CUP · MuJoCo tasks | return and constraint | No clear effect |
| vary cost limit b (e.g. 10 to 30 in Swimmer-v3) (ablation) | CUP · Swimmer-v3 | constraint satisfaction | No clear effect |
Closest papers by meaning
- Safe Reinforcement Learning using Finite-Horizon Gradient-based Estimation (ICML 2024 · similarity 0.90)
- Safe Reinforcement Learning with Instantaneous Constraints: The Role of Aggressive Exploration (AAAI 2024 · similarity 0.88)
- Sampling-Based Safe Reinforcement Learning (NeurIPS 2026 · similarity 0.88)
- Enhancing Efficiency of Safe Reinforcement Learning via Sample Manipulation (NeurIPS 2024 · similarity 0.88)
- Feasibility Consistent Representation Learning for Safe Reinforcement Learning (ICML 2024 · similarity 0.88)
- P2BPO: Permeable Penalty Barrier-Based Policy Optimization for Safe RL (AAAI 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.