Table 1 Statistics of individual task success rate with easy-to-hard task ordering
From: Preserving and combining knowledge in robotic lifelong reinforcement learning
Train after | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Evaluation | Lifelong learning | Multi-task | |||||||||||
Reach | Push | Pick place | Door open | Faucet open | Drawer close | Button press | Peg unplug | Window open | Window close | Forgetting | Forward transfer | ||
Reach | 1.00 | 1.00 | 0.80 | 0.80 | 0.80 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 0.00 | NA | 1.00 |
Push | 0.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 0.80 | 0.80 | 0.80 | 0.20 | 0.00 | 0.80 |
Pick–place | 0.00 | 0.00 | 0.80 | 1.00 | 0.80 | 0.80 | 1.00 | 0.60 | 1.00 | 0.80 | 0.00 | 0.00 | 0.80 |
Door open | 0.00 | 0.00 | 0.00 | 0.40 | 0.80 | 0.80 | 0.60 | 0.40 | 0.80 | 0.60 | −0.20 | 0.00 | 1.00 |
Faucet open | 0.00 | 0.00 | 0.00 | 0.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 0.00 | 0.00 | 1.00 |
Drawer close | 0.00 | 1.00 | 0.80 | 0.80 | 0.00 | 0.60 | 0.80 | 0.80 | 1.00 | 1.00 | −0.40 | 0.52 | 1.00 |
Button press | 0.00 | 0.00 | 0.00 | 0.00 | 0.40 | 0.00 | 0.80 | 0.60 | 0.80 | 0.60 | 0.20 | 0.07 | 1.00 |
Peg unplug | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.20 | 0.00 | 1.00 | 0.60 | 0.60 | 0.40 | 0.03 | 0.80 |
Window open | 0.00 | 0.00 | 0.40 | 0.00 | 0.60 | 0.00 | 0.20 | 0.00 | 0.80 | 1.00 | −0.20 | 0.15 | 1.00 |
Window close | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.40 | 0.00 | 0.40 | 0.40 | 1.00 | NA | 0.13 | 1.00 |
Average | 0.10 | 0.30 | 0.38 | 0.40 | 0.54 | 0.58 | 0.64 | 0.66 | 0.82 | 0.84 | 0.00 | 0.10 | 0.94 |
