发表机构
University of Amsterdam; Linköping University（阿姆斯特丹大学; 林雪平大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对现有机器学习天气预报模型在短时间步长下误差累积和缺乏时间连续性的问题，提出基于神经随机微分方程的SDECast框架，实现连续时间概率预报，并在模拟流和全球小时级预报中验证了其有效性。
AI 中文摘要
现有的机器学习天气预报模型通常通过固定时间分辨率的自回归滚动生成预报。虽然这种方法对于长期预测非常高效，但在使用较短时间步长时可能会遭受严重的误差累积，并且没有明确编码大气动力学的局部性和时间连续性。为了解决这些局限性，我们引入了SDECast，一个用于连续时间概率天气预报的神经随机微分方程（SDE）框架。SDECast扩展了SDE匹配，直接在物理空间中学习随机动力学，无需在训练期间重复进行SDE模拟。在模拟地球物理流上，我们展示了SDECast能够恢复有意义的漂移动力学，并忠实地再现潜在的连续时间行为。然后，我们展示了其在小时间隔分辨率下扩展到全球天气预报的能力，SDECast在长达五天的预测提前期内产生了技巧性的概率预报。
英文摘要
Existing machine learning weather forecasting models typically generate forecasts through autoregressive rollouts at a fixed temporal resolution. While highly efficient for long-range prediction, this formulation can suffer from severe error accumulation when used with shorter time steps and does not explicitly encode the locality and temporal continuity of atmospheric dynamics. To address these limitations, we introduce **SDECast**, a Neural Stochastic Differential Equation (SDE) framework for continuous-time probabilistic weather forecasting. SDECast extends SDE Matching to learn stochastic dynamics directly in physical space, without requiring repeated SDE simulation during training. On a simulated geophysical flow, we show that SDECast recovers meaningful drift dynamics and faithfully reproduces the underlying continuous-time behavior. We then demonstrate its scalability to global weather forecasting at hourly resolution, where SDECast produces skillful probabilistic forecasts for lead times of up to five days.
CommentsAccepted to *AI for Stochastic Dynamics* & *Sim2Science* workshops at NeurIPS 2026
