arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2310.19932cs.LGphysics.ao-ph
收藏

面向环境神经过程的仿真到现实迁移

Sim2Real for Environmental Neural Processes

University of Cambridge(剑桥大学);British Antarctic Survey(英国南极调查局);Vector Institute(向量研究所);Microsoft Research(微软研究院)

Jonas Scholz, Tom R. Andersson, Anna Vaughan, James Requeima, Richard E. Turner

AI总结:

本文研究Sim2Real策略,通过再分析数据预训练ConvCNP并在稀疏气象站观测上微调,显著提升地表气温插值精度,证明再分析数据可作为从真实观测学习的有效起点。

摘要:

基于机器学习(ML)的天气模型近年来取得了快速进展。这些模型通常在来自数值资料同化系统的网格化再分析数据上进行训练。然而,再分析数据存在局限性,例如对物理定律的假设以及较低的时空分辨率。再分析与现实之间的差距引发了人们对直接在气象站等观测数据上训练机器学习模型日益增长的兴趣。对分散且稀疏的环境观测进行建模需要可扩展且灵活的机器学习架构,其中之一是卷积条件神经过程(ConvCNP)。ConvCNP能够学习同时以网格化和非网格化的上下文数据为条件,在目标位置进行具有不确定性感知的预测。然而,真实观测的稀疏性对像ConvCNP这样依赖大量数据的深度学习模型构成了挑战。一种潜在的解决方案是“Sim2Real”:先在再分析数据上进行预训练,再在观测数据上进行微调。我们使用一个经过训练以对德国地表气温进行插值的ConvCNP来分析Sim2Real,并在微调中使用不同数量的气象站。在留出的气象站上,Sim2Real训练显著优于仅使用再分析数据或仅使用气象站数据训练的相同模型架构,表明再分析数据可以作为从真实观测中学习的垫脚石。因此,Sim2Real有望实现更准确的天气预测和气候监测模型。

英文摘要:

Machine learning (ML)-based weather models have recently undergone rapid improvements. These models are typically trained on gridded reanalysis data from numerical data assimilation systems. However, reanalysis data comes with limitations, such as assumptions about physical laws and low spatiotemporal resolution. The gap between reanalysis and reality has sparked growing interest in training ML models directly on observations such as weather stations. Modelling scattered and sparse environmental observations requires scalable and flexible ML architectures, one of which is the convolutional conditional neural process (ConvCNP). ConvCNPs can learn to condition on both gridded and off-the-grid context data to make uncertainty-aware predictions at target locations. However, the sparsity of real observations presents a challenge for data-hungry deep learning models like the ConvCNP. One potential solution is 'Sim2Real': pre-training on reanalysis and fine-tuning on observational data. We analyse Sim2Real with a ConvCNP trained to interpolate surface air temperature over Germany, using varying numbers of weather stations for fine-tuning. On held-out weather stations, Sim2Real training substantially outperforms the same model architecture trained only with reanalysis data or only with station data, showing that reanalysis data can serve as a stepping stone for learning from real observations. Sim2Real could thus enable more accurate models for weather prediction and climate monitoring.

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。

↑