Off-policy prediction on RW tabular
0.023Tail-average RMSETETD
Evaluation Results
| Method | Links | |
|---|---|---|
| TETDalpha=0.01, tail-average window=final 10% of update steps2026.05 | 0.023 | |
| CETDalpha=0.01, tail-average window=final 10% of update steps2026.05 | 0.03 | |
| RETDalpha=0.01, tail-average window=final 10% of update steps2026.05 | 0.037 | |
| ETDalpha=0.01, tail-average window=final 10% of update steps2026.05 | 0.04 | |
| TDalpha=0.01, tail-average window=final 10% of update steps2026.05 | 0.064 | |
| TDRCalpha=0.01, tail-average window=final 10% of update steps2026.05 | 0.069 | |
| RETDalpha=0.01, tail-average window=final 10% of update steps2026.05 | 0.121 | |
| TETDalpha=0.01, tail-average window=final 10% of update steps2026.05 | 0.126 | |
| ETDalpha=0.01, tail-average window=final 10% of update steps2026.05 | 0.129 | |
| CETDalpha=0.01, tail-average window=final 10% of update steps2026.05 | 0.131 | |
| TDalpha=0.01, tail-average window=final 10% of update steps2026.05 | 0.17 | |
| TDRCalpha=0.01, tail-average window=final 10% of update steps2026.05 | 0.175 | |
| GTD2alpha=0.01, tail-average window=final 10% of update steps2026.05 | 0.209 | |
| TDCalpha=0.01, tail-average window=final 10% of update steps2026.05 | 0.215 | |
| TDCalpha=0.01, tail-average window=final 10% of update steps2026.05 | 0.242 | |
| GTD2alpha=0.01, tail-average window=final 10% of update steps2026.05 | 0.272 |