Policy Replication Evaluation on Bricklayer H=49 (replication summary)
0PPO-real Completion RatePPO-real
Evaluation Results
| Method | Links | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| PPO-realH=492026.05 | 0 | 27.8 | — | — | — | — | — | — | — | — | — | |
| Fixed-share DynaH=492026.05 | — | — | — | — | — | 100 | — | 0.523 | — | 0.8 | 0.17 | |
| MDPH=492026.05 | — | — | — | — | — | — | 0.794 | — | — | — | — | |
| Unrestr. DynaH=492026.05 | — | — | 3.1 | 0.028 | 0.009 | — | — | — | — | — | — |