(s, a)-rectangular Robust Policy Gradient Update on Randomly generated l1-robust MDPs
1Relative Running TimePG
Evaluation Results
| Method | Links | |
|---|---|---|
| PGS=10, A=102023.01 | 1 | |
| PGS=30, A=102023.01 | 1 | |
| PGS=50, A=102023.01 | 1 | |
| PGS=100, A=202023.01 | 1 | |
| PGS=500, A=502023.01 | 1 | |
| RPGS=500, A=502023.01 | 1.3 | |
| RPGS=10, A=102023.01 | 1.4 | |
| RPGS=30, A=102023.01 | 1.4 | |
| RPGS=50, A=102023.01 | 1.4 | |
| RPGS=100, A=202023.01 | 1.5 | |
| LP-RPGS=10, A=102023.01 | 326 | |
| LP-RPGS=30, A=102023.01 | 351 | |
| LP-RPGS=50, A=102023.01 | 408 | |
| LP-RPGS=100, A=202023.01 | 469 | |
| LP-RPGS=500, A=502023.01 | 925 |