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

Doorkey

Benchmarks

Task NameDataset NameSOTA ResultTrend
NavigationDoorKey 8x8 100 episodes (test)
Reward91
7
Reinforcement LearningDoorKey
Reward0.97
7
Reinforcement LearningDoorKey 6x6 (train)
Mean Episodic Return0.744
5
Extrapolation performanceDoorKey (test)
Accuracy100
4
MDP AbstractionDoorKey
Compression Rate100
3
ExtrapolationDoorKey 256x256 (test)
Accuracy100
2
ExtrapolationDoorkey
Accuracy100
2
Showing 7 of 7 rows