Offline Reinforcement Learning on AntMaze-Ultra-Play benchmark leaderboard · SOTA2 Research

Offline Reinforcement Learning on AntMaze-Ultra-Play

56.6Avg Normalized ScoreGCPC

Evaluation Results

Method	Links
GCPC2023.07		56.6
RvS-G2023.07		25.6
DWSL2023.07		25.2
TAP2023.07		22
IQL2023.07		21.2
GCIQL2023.07		20.6
CQL2023.07		18
VIPO-LEQ2025.04		0.31
LEQ2025.04		0.258
IQL-TD-MPC2025.04		0.206
WGCSL2023.07		0.2
BC2023.07		0
DT2023.07		0
MOBILE2025.04		0