Overcoming "Physics Shock" in Earth Observation A Heteroscedastic Uncertainty Framework for PINN-based Flood Inference

In three linesHeteroscedastic uncertainty-aware PINN framework for flood extent mapping from SAR data. Attention-Gated FNO-UNet with dynamic Warm-Start protocol and aleatoric uncertainty modeling prevents gradient divergence ("Physics Shock"). On Sen1Floods11: +25% relative IoU improvement over deterministic baselines, with calibrated confidence bounds for disaster response.

Summary generated by Claude — human-verified