Reinforcement Learning on MuJoCo Halfcheetah epsilon=0.15 (test) benchmark leaderboard · SOTA2 Research

Reinforcement Learning on MuJoCo Halfcheetah epsilon=0.15 (test)

5,048Natural ReturnPPO

Evaluation Results

Method	Links
PPO2025.12		5,048	4,463	3,281	918	1,049	-213	-69	-213	-1.042
BAR2025.12		4,837	4,741	4,803	4,729	4,265	3,894	3,181	3,181	-0.342
SA2025.12		4,780	4,983	5,035	3,759	2,727	1,443	1,551	1,443	-0.698
RADIAL2025.12		4,739	4,642	4,546	2,961	1,327	1,522	1,968	1,327	-0.72
WocaR2025.12		4,723	4,798	4,846	4,543	3,302	2,270	2,498	2,270	-0.519