Reinforcement Learning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
4Episode Length
2
Feb 26, 2026
3Length
2
Feb 26, 2026
25.8Score
2
Feb 26, 2026
5,386Ant Score
2
Feb 26, 2026
279,703.5Alien Score
2
Feb 26, 2026
0.7Maximum Average Return
2
Feb 26, 2026
3Max Average Return
2
Feb 26, 2026
207Median Human-Normalized Score
2
Feb 26, 2026
378.6Initial Score (Offline RL)
2
Feb 26, 2026
634,972,384Score
2
Feb 26, 2026
51,022Score
2
Feb 26, 2026
2,950.1Alien Score
2
Feb 26, 2026
15,229Assault Score
2
Feb 26, 2026
2,152Alien Score
2
Feb 26, 2026
78.9P(Goal)
2
Feb 26, 2026
98.7Return
2
Feb 26, 2026
5Score 100
2
Feb 26, 2026
98Best Stable Score
2
Feb 26, 2026
1,350Average Score
2
Feb 26, 2026
192Average Score
2
Feb 26, 2026
6.08Time
1
Mar 23, 2026
16.43Time
1
Mar 23, 2026
0.03Time
1
Mar 23, 2026
0.62Time
1
Mar 23, 2026
2.97Time
1
Mar 23, 2026