Representative robustness analysis under noise and physics perturbations
| Test case | Test RMSE (0–180 days) | Observed long-horizon behavior |
|---|---|---|
| Baseline training configuration | 0.008 | Smooth and monotonic degradation trajectory |
| Increased noise (≈2 × baseline) | 0.010 | Stable long-horizon degradation behavior maintained |
| Increased noise (≈3 × baseline) | 0.013 | Smooth extrapolation without oscillatory divergence |
| Diffusion coefficient perturbed (±20%) | 0.015 | No numerical instability or non-physical divergence observed |
| Test case | Test RMSE (0–180 days) | Observed long-horizon behavior |
|---|---|---|
| Baseline training configuration | 0.008 | Smooth and monotonic degradation trajectory |
| Increased noise (≈2 × baseline) | 0.010 | Stable long-horizon degradation behavior maintained |
| Increased noise (≈3 × baseline) | 0.013 | Smooth extrapolation without oscillatory divergence |
| Diffusion coefficient perturbed (±20%) | 0.015 | No numerical instability or non-physical divergence observed |
Sharing content requires targeting cookies to be enabled. Please update your cookie preferences to use this feature.