arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.06800cs.LG
收藏

FlowDA:通过流匹配实现准确且低延迟的天气数据同化

FlowDA: Accurate, Low-Latency Weather Data Assimilation via Flow Matching

National University of Singapore(新加坡国立大学)

Ran Cheng, Lailai Zhu

AI总结:

FlowDA通过流匹配实现低延迟、高精度的天气数据同化,展现出在不同观测率下的优越性能和鲁棒性。

摘要:

数据同化(DA)是现代天气预测中的基本组成部分,然而,由于依赖于传统变分方法,它仍然是基于机器学习(ML)的预测流水线中的主要计算瓶颈。最近的生成ML基于DA方法提供了一种有希望的替代方案,但通常需要许多采样步骤,并且在长跨度自回归滚动中使用循环同化时会累积误差。我们提出了FlowDA,一种基于流匹配的低延迟天气尺度生成DA框架。FlowDA通过基于SetConv的嵌入对观测进行条件,并微调Aurora基础模型以提供准确、高效且稳健的分析。在观测率从3.9%降至0.1%的实验中,FlowDA在与类似可调参数大小的强基线相比表现更优。FlowDA进一步显示出对观测噪声的鲁棒性和在长跨度自回归循环DA中的稳定性能。总体而言,FlowDA指出了数据驱动DA的高效且可扩展的方向。

英文摘要:

Data assimilation (DA) is a fundamental component of modern weather prediction, yet it remains a major computational bottleneck in machine learning (ML)-based forecasting pipelines due to reliance on traditional variational methods. Recent generative ML-based DA methods offer a promising alternative but typically require many sampling steps and suffer from error accumulation under long-horizon auto-regressive rollouts with cycling assimilation. We propose FlowDA, a low-latency weather-scale generative DA framework based on flow matching. FlowDA conditions on observations through a SetConv-based embedding and fine-tunes the Aurora foundation model to deliver accurate, efficient, and robust analyses. Experiments across observation rates decreasing from $3.9\%$ to $0.1\%$ demonstrate superior performance of FlowDA over strong baselines with similar tunable-parameter size. FlowDA further shows robustness to observational noise and stable performance in long-horizon auto-regressive cycling DA. Overall, FlowDA points to an efficient and scalable direction for data-driven DA.

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。

↑