Tyche：一步流用于高效的概率天气预报
Tyche: One Step Flow for Efficient Probabilistic Weather Forecasting
AI总结：
Tyche提出一种一步条件流模型，通过JVP正则化目标和Swin式变换器，在高维地理场中实现高效概率天气预报，实验显示其性能优于多步生成基线和ECMWF IFS集合预报。
摘要：
概率天气预报不仅需要准确的轨迹，还需要在合理的大气未来上校准分布。最近的数据驱动系统在确定性技能上取得了显著进展，扩散基的集合预报器显著提高了样本真实性和不确定性量化。然而，其推理成本随预报时间、集合大小和每个转换所需的去噪步骤数量而增加，使大规模操作集合昂贵。为此，我们提出了Tyche，一种一步条件流模型，用于高效的概率天气预报。Tyche通过目的地感知的平均速度流模型，将高斯噪声直接映射到未来天气状态，单次函数评估（1-NFE）。为了使这种一步传输在高维地理场中可学习，我们推导了一个JVP正则化的校正目标，强制源和目的地流时间步之间的时间自一致性，而无需显式形成雅可比矩阵。传输场由一个各向同性的Swin式变换器参数化，保留了细尺度空间结构，同时在全球网格上保持可扩展性。为了提高自回归预报下的集合可靠性，我们进一步引入了基于回放的微调阶段，结合课程CRPS校准监督。在ERA5数据上1.5°和6小时分辨率的实验显示，我们的Tyche仅使用一次NFE，其预报技能和校准性能与最先进的多步生成基线和操作ECMWF IFS集合相当。
英文摘要：
Probabilistic weather forecasting requires not only accurate trajectories, but calibrated distributions over plausible atmospheric futures. Recent data-driven systems have achieved remarkable deterministic skill, and diffusion-based ensemble forecasters have substantially improved sample realism and uncertainty quantification. However, their inference cost scales with forecast horizon, ensemble size, and the number of denoising steps required for each transition, making large operational ensembles expensive. To address this, we present Tyche, a one-step conditional flow model for efficient probabilistic weather forecasting. Tyche models the conditional forecast distribution with a destination-aware average-velocity flow that maps Gaussian noise directly to future weather states in a single function evaluation (1-NFE). To make this one-step transport learnable in high-dimensional geophysical fields, we derive a JVP-regularized rectification objective that enforces temporal self-consistency across source and destination flow timesteps without explicitly forming Jacobians. The transport field is parameterized by an isotropic Swin-style transformer that preserves fine-scale spatial structure while remaining scalable on global grids. To improve ensemble reliability under autoregressive forecasting, we further introduce a rollout-based finetuning stage with curriculum CRPS calibration supervision. Experiments on ERA5 at 1.5$^\circ$ and 6-hour resolution show that our Tyche, using merely a single NFE, matches or exceeds the forecast skill and calibration of state-of-the-art multi-step generative baselines and the operational ECMWF IFS ensemble.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
