Policy Evaluation on BPIC real logs 2012
374.966Average KPI (Mπ)MDP-based RL method
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| MDP-based RL methodvariant=sigQ12023.03 | 374.966 | 886.098 | |
| Offline Deep RL method (CQL)iterations=1002023.03 | 373.028 | 854.281 | |
| Offline Deep RL method (CQL)iterations=502023.03 | 369.776 | 851.857 | |
| MDP-based RL methodvariant=stepQ12023.03 | 368.507 | 914.636 | |
| MDP-based RL methodvariant=stepQ22023.03 | 366.831 | 864.043 | |
| customary policy2023.03 | 274.78 | 688.403 |