D4RL
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
D4RL halfcheetah-mr
96.3Final Online Score
6
D4RL walker2d-m
118.3Final Online Score
6
D4RL hopper-m
104.3Final Online Score
6
D4RL halfcheetah-m
99.6Final Online Score
6
D4RL FrankaKitchen
58.1D4RL Normalized Score
6
D4RL adroit (8 tasks)
71Normalized Return
6
D4RL (held-out evaluator)
2.97Temporal Error
6
D4RL Walker2d
1,526Return
6
D4RL Hopper
892Return
6
D4RL HalfCheetah
3,540Return
6
D4RL
3,275Hopper Score
6
D4RL Locomotion
3,275Hopper-v2 Score
6
D4RL Kitchen Tasks
79.3Kitchen Complete v0 Success
6
D4RL Manipulation v1
117Normalized Score
6
D4RL MuJoCo locomotion v2
0.005Runtime (s)
6
D4RL AntMaze v0 (online fine-tuning)
96.9Success Rate (umaze-v0)
6
D4RL Locomotion online fine-tuning v2
97.6Hopper v2 MR Score
6
D4RL POMDP HalfCheetah Medium
47.1Normalized Score
6
D4RL POMDP HalfCheetah (Medium-Expert)
94.9Normalized Score
6
D4RL Halfcheetah-medium-expert v0
7,989Return
6
D4RL Walker2d medium-expert v0
1,515Return
6
D4RL Hopper-medium-expert v0
3,907Return
6
D4RL Hopper-medium v0
1,676Return
6
D4RL Hopper-random v0
685Return
6
D4RL walker2d-medium v0
1,269Return
6