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

K2D9

Benchmarks

Task NameDataset NameSOTA ResultTrend
Offline Reinforcement LearningK2D9 1000-5-mid
NAUC95
4
Offline Reinforcement LearningK2D9-1000-1 mid
NAUC93
4
Offline Reinforcement LearningK2D9-500-5 mid
NAUC92
4
Offline Reinforcement LearningK2D9 mid 500-1
NAUC89
4
Offline Reinforcement LearningK2D9 250-5 mid
NAUC82
4
Offline Reinforcement LearningK2D9 250-1-mid
NAUC78
4
Reinforcement Learning target performance estimationK2D9 1000-5-late
NAUC0.86
4
Reinforcement Learning target performance estimationK2D9 late 1000-1
NAUC80
4
Reinforcement Learning target performance estimationK2D9 500-5-late
NAUC83
4
Reinforcement Learning target performance estimationK2D9-500-1-late
NAUC75
4
Reinforcement Learning target performance estimationK2D9 250-5-late
NAUC74
4
Reinforcement Learning target performance estimationK2D9 250-1-late
NAUC39
4
Offline In-Context Reinforcement LearningK2D9-1000-5 (complete)
NAUC96
4
Offline In-Context Reinforcement LearningK2D9-500-5 (complete)
NAUC94
4
Offline In-Context Reinforcement LearningK2D9-1000-5-early (test)
NAUC85
4
Offline In-Context Reinforcement LearningK2D9-1000-1-early (test)
NAUC0.67
4
Offline In-Context Reinforcement LearningK2D9-500-5-early (test)
NAUC0.74
4
Offline In-Context Reinforcement LearningK2D9-250-5-early (test)
NAUC58
4
Offline In-Context Reinforcement LearningK2D9-250-1-early (test)
NAUC38
4
Showing 19 of 19 rows