Offline Reinforcement Learning on D4RL antmaze-umaze (fixed) benchmark leaderboard · SOTA2 Research

Offline Reinforcement Learning on D4RL antmaze-umaze (fixed)

82Normalized ScoreBC

Evaluation Results

Method	Links
BC2021.10		82
COMBO2021.10		80.3
BCQ2021.10		79.5
ROMI-BCQ2021.10		68.7
ROMI-CQL2021.10		68.1
CQL2021.10		66.9
BAIL2021.10		58
BEAR2021.10		0
BRAC-P2021.10		0
BRAC-v2021.10		0
MOPO2021.10		0
MOREL2021.10		0
Repb-SDE2021.10		0