spliif
Sparse Local Implicit Image Function for sub-km Weather Downscaling
Redondo, Yago del Valle Inclan, Arriaga-Varela, Enrique, Lyamzin, Dmitry, Cervantes, Pablo, Ramalho, Tiago
We introduce SpLIIF to generate implicit neural representations and enable arbitrary downscaling of weather variables. We train a model from sparse weather stations and topography over Japan and evaluate in- and out-of-distribution accuracy predicting temperature and wind, comparing it to both an interpolation baseline and CorrDiff. We find the model to be up to 50% better than both CorrDiff and the baseline at downscaling temperature, and around 10-20% better for wind.
2510.20228
Technology: Information Technology > Artificial Intelligence > Machine Learning > Neural Networks > Deep Learning (0.69)