Table 6 Sensitivity to the robust trigger hyperparameter gamma across multiple datasets with baseline references. Baseline and EGMF-GR serve as reference anchors under identical settings for each dataset. Best gamma reports the lowest MSE on the tested grid.
From: Hybrid evolutionary-gradient training improves long-term time series forecasting
Dataset | Baseline | EGMF-GR | Baseline | EGMF-GR | Best gamma |
|---|---|---|---|---|---|
| Â | MSE MAE | MSE MAE | MSE | MSE | Trigger |
ETTm1 | 0.349370 0.375490 | 0.323102 0.364091 | 0.349370 | 0.323102 | 0.0 1.0000 |
ETTm2 | 0.185490 0.264592 | 0.177144 0.262366 | 0.185490 | 0.177144 | 0.0 1.0000 |
ETTh1 | 0.410940 0.419820 | 0.394974 0.407395 | 0.410940 | 0.394974 | 0.0 1.0000 |
ETTh2 | 0.336645 0.361489 | 0.298155 0.347697 | 0.336645 | 0.298155 | 1.0 0.4000 |
