Data-efficient flood depth prediction through domain-aware coreset selection and tabular foundation models
arXiv:2606.05265v1 Announce Type: new Abstract: Near-real-time flood depth prediction demands surrogate models that are accurate, fast, and transferable across watersheds. Supervised surrogates can match physics-based simulators in accuracy but need millions of training rows per watershed and cannot extrapolate beyond their original mesh. We propose a domain-aware coreset construction pipeline that conditions a tabular foundation model at inference time. The pipeline stratifies storms by return period and most-affected watershed, then samples hexagons with a target-aware spatial selector. With
The increasing frequency and intensity of extreme weather events necessitate more robust and data-efficient prediction models; AI advancements are now enabling this scalability.
Accurate, real-time flood prediction is critical for disaster preparedness, infrastructure planning, and mitigating economic losses associated with climate change impacts, directly affecting resource-intensive sectors.
Flood depth prediction can now be achieved with significantly less training data per watershed, enabling faster deployment and broader application of high-accuracy models across diverse geographies.
- · Emergency services
- · Insurance companies
- · Urban planners
- · Climate resilience tech
- · Traditional flood modeling services reliant on extensive data
- · Regions without access to advanced AI infrastructure
- · Communities unprepared for climate impacts
Improved flood warnings and targeted mitigation efforts become possible, reducing immediate flood damages.
More reliable flood risk assessments enable better land use planning and more accurate insurance pricing, influencing real estate markets and development.
Enhanced predictive capabilities for water-related hazards could inform broader strategies for water resource management and climate adaptation, potentially easing pressures on water-stressed regions.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.LG