Policy Value Estimation on Scenario 3
1.879Policy Value (mean)PDRO-ITR
Evaluation results
| Method | Links | ||
|---|---|---|---|
| PDRO-ITRSample size (n)=2000, delta=0.752026.03 | 1.879 | 0.015 | |
| PDRO-ITRSample size (n)=1000, delta=0.752026.03 | 1.826 | 0.024 | |
| PDRO-ITRSample size (n)=500, delta=0.752026.03 | 1.607 | 0.149 | |
| NaiveSample size (n)=2000, delta=0.752026.03 | 1.357 | 0.032 | |
| DRO-ITRSample size (n)=2000, delta=0.752026.03 | 1.354 | 0.041 | |
| NaiveSample size (n)=1000, delta=0.752026.03 | 1.331 | 0.047 | |
| MPLSample size (n)=2000, delta=0.752026.03 | 1.316 | 0.035 | |
| MPLSample size (n)=1000, delta=0.752026.03 | 1.307 | 0.045 | |
| DRO-ITRSample size (n)=1000, delta=0.752026.03 | 1.293 | 0.094 | |
| MPLSample size (n)=500, delta=0.752026.03 | 1.256 | 0.069 | |
| NaiveSample size (n)=500, delta=0.752026.03 | 1.176 | 0.133 | |
| MR-CATESample size (n)=2000, delta=0.752026.03 | 1.027 | 0.035 | |
| DRO-ITRSample size (n)=500, delta=0.752026.03 | 1.024 | 0.287 | |
| MR-CATESample size (n)=1000, delta=0.752026.03 | 1.015 | 0.045 | |
| MR-CATESample size (n)=500, delta=0.752026.03 | 0.949 | 0.092 |