Reinforcement Learning on Safe CartPole benchmark leaderboard · SOTA2 Research

Reinforcement Learning on Safe CartPole

200Episode RewardRPO-DDPG

Evaluation Results

Method	Links
RPO-DDPG2023.10		200	0	0	0	0
RPO-SAC2023.10		200	0	0	0	0
CUP2023.10		63.9	0	0.5344	0.7904	0
Safety Layer2023.10		53.8	0.0086	1.7099	0	13.6603
SAC-L2023.10		40.5	0	0.0487	0	0.1038
CPO2023.10		20.1	0	0.0389	0	0.1096
DDPG-L2023.10		15.5	0	0.0408	0	0.118