Offline Reinforcement Learning on D4RL maze2d-umaze sparse benchmark leaderboard · SOTA2 Research

Offline Reinforcement Learning on D4RL maze2d-umaze sparse

139.5Normalized ScoreROMI-BCQ

Evaluation Results

Method	Links
ROMI-BCQ2021.10		139.5
COMBO2021.10		76.4
BEAR2021.10		65.7
ROMI-CQL2021.10		54.3
BCQ2021.10		49.1
BAIL2021.10		44.7
CQL2021.10		18.9
BRAC-v2021.10		8.7
BC2021.10		-3.2
BRAC-P2021.10		-9.2
Repb-SDE2021.10		-12.3
MOREL2021.10		-14.4
MOPO2021.10		-14.8