English

A multi-scale loss formulation for learning a probabilistic model with proper score optimisation

Atmospheric and Oceanic Physics 2025-06-13 v1 Artificial Intelligence

Abstract

We assess the impact of a multi-scale loss formulation for training probabilistic machine-learned weather forecasting models. The multi-scale loss is tested in AIFS-CRPS, a machine-learned weather forecasting model developed at the European Centre for Medium-Range Weather Forecasts (ECMWF). AIFS-CRPS is trained by directly optimising the almost fair continuous ranked probability score (afCRPS). The multi-scale loss better constrains small scale variability without negatively impacting forecast skill. This opens up promising directions for future work in scale-aware model training.

Keywords

Cite

@article{arxiv.2506.10868,
  title  = {A multi-scale loss formulation for learning a probabilistic model with proper score optimisation},
  author = {Simon Lang and Martin Leutbecher and Pedro Maciel},
  journal= {arXiv preprint arXiv:2506.10868},
  year   = {2025}
}
R2 v1 2026-07-01T03:13:50.027Z