Reinforcement Learning on Maze 17^10 unstructured discrete
9.25Mean ScoreDGRL (SDN, DBU)
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| DGRL (SDN, DBU)Num Seeds=102026.02 | 9.25 | 9.29 | 0.18 | 66 | 2,290 | |
| DNC (SA), A2CNum Seeds=102026.02 | 5.5 | 5.59 | 3.28 | 0 | 1,338 | |
| DNC (greedy), A2CNum Seeds=102026.02 | 5.26 | 4.86 | 3.27 | -13 | 1,150 | |
| Cacla, A2CNum Seeds=102026.02 | 1.07 | 0.39 | 1.47 | -93 | 0 |