Research map / Reinforcement learning / Visual and representation RL
OmniSafe: An Infrastructure for Accelerating Safe Reinforcement Learning Research
Jiaming Ji, Jiayi Zhou, Borong Zhang, Juntao Dai, Xuehai Pan, Ruiyang Sun, Weidong Huang, Yiran Geng, Mickel Liu, Yaodong Yang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 OmniSafe implementation instead of Tianshou | PPO · SafetyAntVelocity-v1 | episode reward: 2607.48 -> 4295.96 | Improved |
| use OmniSafe implementation instead of Stable-Baselines3 | PPO · SafetyAntVelocity-v1 | episode reward: 1780.61 -> 4295.96 | Improved |
| use OmniSafe implementation instead of Tianshou | PPO · SafetyHalfCheetahVelocity-v1 | episode reward: 6299.27 -> 3507.47 | Hurt |
| use OmniSafe implementation instead of Stable-Baselines3 | PPO · SafetySwimmerVelocity-v1 | episode reward: 273.86 -> 113.28 | Hurt |
| use OmniSafe implementation instead of Tianshou | TRPO · SafetyHopperVelocity-v1 | episode reward: 1346.94 -> 2698.19 | Improved |
| use OmniSafe implementation instead of Stable-Baselines3 | TRPO · SafetyHumanoidVelocity-v1 | episode reward: 2828.18 -> 1461.51 | Hurt |
| use OmniSafe implementation instead of Stable-Baselines3 | SAC · SafetyAntVelocity-v1 | episode reward: 2404.50 -> 5456.31 | Improved |
| use OmniSafe implementation instead of Tianshou | SAC · SafetyAntVelocity-v1 | episode reward: 6012.30 -> 5456.31 | Hurt |
Closest papers by meaning
- SafeDreamer: Safe Reinforcement Learning with World Models (ICLR 2024 · similarity 0.89)
- Balance Reward and Safety Optimization for Safe Reinforcement Learning: A Perspective of Gradient Manipulation (AAAI 2024 · similarity 0.88)
- Explainably Safe Reinforcement Learning (NeurIPS 2025 · similarity 0.87)
- Offline Safe Reinforcement Learning Using Trajectory Classification (AAAI 2025 · similarity 0.87)
- Enhancing Efficiency of Safe Reinforcement Learning via Sample Manipulation (NeurIPS 2024 · similarity 0.87)
- A Provable Approach for End-to-End Safe Reinforcement Learning (NeurIPS 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.
