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 2409.04779cs.LGcs.NAmath.NA
收藏

用于奇异摄动微分方程的分量傅里叶神经算子

Component Fourier Neural Operator for Singularly Perturbed Differential Equations

Nanjing University of Aeronautics and Astronautics(南京航空航天大学);Tsinghua University(清华大学)

Ye Li, Ting Du, Yiwen Pang, Zhongyi Huang

AI总结:

本文提出融合渐近分析先验的 ComFNO 算子学习方法,用于求解含薄层快速转变的奇异摄动微分方程,并在多类问题上显著提升 FNO 的精度与少样本泛化能力。

摘要:

求解奇异摄动微分方程(SPDEs)面临计算挑战,这些挑战源于其解在薄区域内的快速转变。深度学习在处理微分方程方面的有效性,促使我们采用这些方法来求解 SPDEs。在本手稿中,我们提出了 Component Fourier Neural Operator(ComFNO,分量傅里叶神经算子),这是一种建立在 Fourier Neural Operator(FNO,傅里叶神经算子)基础上的创新算子学习方法,同时融入了从渐近分析中获得的有价值先验知识。我们的方法并不局限于 FNO,还可应用于其他神经网络框架,例如 Deep Operator Network(DeepONet,深度算子网络),从而产生潜在的类似 SPDEs 求解器。跨多种 SPDEs 类别的实验结果表明,与原始 FNO 相比,ComFNO 显著提高了精度。此外,ComFNO 对不同数据分布表现出天然的适应性,并在少样本场景中表现良好,展示了其在实际情形中出色的泛化能力。

英文摘要:

Solving Singularly Perturbed Differential Equations (SPDEs) poses computational challenges arising from the rapid transitions in their solutions within thin regions. The effectiveness of deep learning in addressing differential equations motivates us to employ these methods for solving SPDEs. In this manuscript, we introduce Component Fourier Neural Operator (ComFNO), an innovative operator learning method that builds upon Fourier Neural Operator (FNO), while simultaneously incorporating valuable prior knowledge obtained from asymptotic analysis. Our approach is not limited to FNO and can be applied to other neural network frameworks, such as Deep Operator Network (DeepONet), leading to potential similar SPDEs solvers. Experimental results across diverse classes of SPDEs demonstrate that ComFNO significantly improves accuracy compared to vanilla FNO. Furthermore, ComFNO exhibits natural adaptability to diverse data distributions and performs well in few-shot scenarios, showcasing its excellent generalization ability in practical situations.

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。

↑