Robotic Manipulation on Adroit pen, door, relocate, hammer v0
56.9Pen Score (Cloned)BC
Evaluation Results
| Method | Links | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| BCrandom_seeds=5, normalization=0: random policy, 100: expert policy2022.06 | 56.9 | 34.4 | 85.1 | -0.1 | 0.5 | 34.9 | -0.1 | 0 | 101.3 | 0.8 | 1.5 | 125.6 | 440.8 | |
| CABI+BCQrandom_seeds=5, normalization=0: random policy, 100: expert policy, averaging_protocol=final 10 evaluations2022.06 | 54.7 | 75.1 | 127.6 | 0.5 | 1.7 | 105.3 | -0.2 | 0.1 | 105.9 | 4.3 | 3.1 | 128.9 | 607 | |
| BCQrandom_seeds=5, normalization=0: random policy, 100: expert policy2022.06 | 44 | 68.9 | 114.9 | 0 | 0 | 99 | -0.3 | 0.5 | 41.6 | 0.4 | 0.5 | 107.2 | 476.7 | |
| CQLrandom_seeds=5, normalization=0: random policy, 100: expert policy2022.06 | 39.2 | 37.5 | 107 | 0.4 | 9.9 | 101.5 | -0.1 | 0.2 | 95 | 2.1 | 4.4 | 86.7 | 483.8 | |
| UWACrandom_seeds=5, normalization=0: random policy, 100: expert policy2022.06 | 33.1 | 21.7 | 111.9 | 0 | 2.1 | 104.1 | -0.3 | 0.5 | 105.6 | 0.4 | 1.1 | 110.6 | 490.8 | |
| AWRrandom_seeds=5, normalization=0: random policy, 100: expert policy2022.06 | 28 | 12.3 | 111 | 0 | 0.4 | 102.9 | -0.2 | 0 | 91.5 | 0.4 | 1.2 | 39 | 386.5 | |
| BEARrandom_seeds=5, normalization=0: random policy, 100: expert policy2022.06 | 26.5 | -1 | 105.9 | -0.1 | -0.3 | 103.4 | -0.3 | -0.3 | 98.6 | 0.3 | 0.3 | 127.3 | 460.3 | |
| MOPOrandom_seeds=5, normalization=0: random policy, 100: expert policy2022.06 | -2.1 | 9.7 | -0.6 | -0.1 | -0.2 | -0.2 | -0.3 | -0.3 | -0.2 | 0.2 | 0.2 | 0.3 | 6.4 | |
| COMBOrandom_seeds=5, normalization=0: random policy, 100: expert policy2022.06 | -2.4 | 27.7 | 11.5 | 0 | -0.3 | 4.9 | -0.1 | -0.3 | 17.2 | 0.4 | 0.2 | 0.3 | 59.1 |