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

Reinforcement Learning on MuJoCo Ant epsilon=0.15 (test)

5,381Natural ReturnPPO

Evaluation Results

Method	Links
PPO2025.12		5,381	5,329	4,696	1,768	1,097	-1,398	-3,107	-3,107	-1.577
SA2025.12		5,367	5,217	5,012	5,114	4,396	4,227	2,355	2,355	-0.539
BAR2025.12		5,024	4,979	4,777	4,843	4,171	3,367	2,825	2,825	-0.438
RADIAL2025.12		4,358	4,309	3,628	4,205	3,742	2,364	3,261	2,364	-0.462
WocaR2025.12		4,069	3,911	3,978	3,689	3,176	1,868	1,830	1,830	-0.55