Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Offline Reinforcement Learning on AntMaze umaze-diverse (Normalized Average Return)

84Normalized Average Return

CQL

-3.3619.324264.68Apr 16, 2025Jun 22, 2025Aug 29, 2025Nov 5, 2025Jan 11, 2026Mar 20, 2026May 27, 2026
Updated 10d ago

Evaluation Results

MethodLinks
2026.05
84
2026.05
83.7
2026.05
78.5
2025.04
74.8
2026.05
73.2
2025.04
72.6
2026.05
72.3
2026.05
71.4
2025.04
71
2026.05
71
2026.05
70.1
2026.05
62.2
2026.05
51.2
2025.04
0