Offline Reinforcement Learning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
92NAUC
4
May 27, 2026
89NAUC
4
May 27, 2026
82NAUC
4
May 27, 2026
78NAUC
4
May 27, 2026
85NAUC
4
May 27, 2026
43NAUC
4
May 27, 2026
57NAUC
4
May 27, 2026
27NAUC
4
May 27, 2026
34NAUC
4
May 27, 2026
17NAUC
4
May 27, 2026
77NAUC
4
May 27, 2026
47NAUC
4
May 27, 2026
51NAUC
4
May 27, 2026
0.37NAUC
4
May 27, 2026
0.34NAUC
4
May 27, 2026
27NAUC
4
May 27, 2026
-3Sample Complexity
4
May 19, 2026
95.7Normalized Average Return
4
May 14, 2026
70.8Ant-m (m)
4
May 14, 2026
33.74HalfCheetah Score (Observation)
4
Apr 8, 2026
38.46HalfCheetah Score (Observation)
4
Apr 8, 2026
132.3U-Maze Score
4
Mar 17, 2026
118.4Normalized Score
4
Feb 26, 2026
112.3Normalized Score
4
Feb 26, 2026
95.8Normalized Score
4
Feb 26, 2026