Offline Reinforcement Learning on D4RL Antmaze (Play)
88.8AntMaze Medium ScoreDiffuserLite
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| DiffuserLiteBackbone=R2, Runtime per action (second)=0.0082024.01 | 88.8 | 69.4 | 8,030 | |
| DiffuserLiteBackbone=R1, Runtime per action (second)=0.0142024.01 | 88 | 72.4 | 8,250 | |
| DiffuserLiteBackbone=D, Runtime per action (second)=0.022024.01 | 78 | 72 | 7,760 | |
| CQL2024.01 | 65.8 | 20.8 | 4,360 | |
| IQL2024.01 | 65.8 | 42 | 5,300 | |
| BC2024.01 | 0 | 0 | 20 | |
| DiffuserRuntime per action (second)=0.7492024.01 | 0 | 0 | 0 | |
| DDRuntime per action (second)=2.4352024.01 | 0 | 0 | 0 |