Weather extremes pose major societal risks, especially in a changing climate, but due to their rarity, they are difficult to study using limited observations or complex climate models. We introduce AI+RES, a framework coupling fast AI weather forecasts with a high-fidelity physics model using a rare-event algorithm to efficiently characterize extremes. This approach enables the study of the statistics and physics of very rare events, such as once per millennium heatwaves at two orders-of-magnitude lower computational cost. AI+RES can be applied broadly across climate science and other fields concerned with rare events.
@article{arxiv.2510.27066,
title = {AI-boosted rare event sampling to characterize extreme weather},
author = {Amaury Lancelin and Alex Wikner and Laurent Dubus and Clément Le Priol and Dorian S. Abbot and Freddy Bouchet and Pedram Hassanzadeh and Jonathan Weare},
journal= {arXiv preprint arXiv:2510.27066},
year = {2026}
}