arXiv：cs.LG· Maria Marchenko, Martin Andrae, Fredrik Lindsten, Christian A. Naesseth·· 3 小时前AI 评分34
SDECast：基于神经 SDE 的连续时间概率天气预报
SDECast: Probabilistic Weather Forecasting in Continuous Time with Neural SDEs
AI 导读
SDECast 是一个基于神经随机微分方程（SDE）的连续时间概率天气预报框架，通过扩展 SDE Matching 直接在物理空间学习随机动力学，训练时无需重复进行 SDE 模拟。
正文
Abstract: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.
| Comments: | Accepted to *AI for Stochastic Dynamics* & *Sim2Science* workshops at NeurIPS 2026 |
| Subjects: | Machine Learning (stat.ML); Machine Learning (cs.LG); Atmospheric and Oceanic Physics (physics.ao-ph) |
| Cite as: | arXiv:2610.03313 [stat.ML] |
| (or arXiv:2610.03313v1 [stat.ML] for this version) | |
| https://doi.org/10.48550/arXiv.2610.03313 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Martin Andrae [view email]
[v1]
Fri, 2 Oct 2026 13:50:59 UTC (5,570 KB)
来源：arXiv：cs.LG · arxiv.org