Skip to content

Latest commit

 

History

History
32 lines (24 loc) · 1.3 KB

README.md

File metadata and controls

32 lines (24 loc) · 1.3 KB

Safety MuJoCo

We are currently designing safety tasks for safe RL based on MuJoCo environments and implementing more algorithms in the benchmark.

Figure.1 Safety-MuJoCo Environments: SafetyWalker-4 (a), SafetyHumanoidStandup-v4 (b), SafetyReacher-v4 (c), SafetyHopper-v4 (d), SafetyAnt-v4 (e), SafeHalfCheetah-v4 (f), SafetyPusher-v4 (g) and SafetyHumanoid-v4 (h).

Publication

If you find the repository useful, please cite the paper:

@inproceedings{gu2024balance,
  title={Balance Reward and Safety Optimization for Safe Reinforcement Learning: A Perspective of Gradient Manipulation},
  author={Gu, Shangding and Sel, Bilgehan and Ding, Yuhao and Wang, Lu and Lin, Qingwei and Jin, Ming and Knoll, Alois},
  booktitle={Proceedings of the AAAI Conference on Artificial Intelligence},
  volume={38},
  number={19},
  pages={21099--21106},
  year={2024}
}