arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2310.11271math.NAcs.NA
收藏

混合有限元/神经网络离散化的误差分析

Error analysis for hybrid finite element/neural network discretizations

Otto-von-Guericke-Univeristät Magdeburg(奥托·冯·古里克马格德堡大学);Universität der Bundeswehr München(慕尼黑联邦国防军大学)

Uladzislau Kapustsin, Utku Kaya, Johannes Pfefferer, Thomas Richter

AI总结:

本文提出一种混合有限元与神经网络的方法,通过局部修正粗解以获得精细解,并分析其误差与训练集大小的关系,验证了方法在不同域上的泛化能力。

摘要:

我们描述并分析了一种混合有限元/神经网络方法,用于预测偏微分方程的解。该方法设计用于以局部方式从神经网络中获取细尺度波动。网络能够根据源项和粗略有限元近似作为输入,局部地将粗略有限元解修正为精细解。关键观察是预测质量与训练集大小之间的依赖关系,训练集包含不同源项及其对应的精细和粗略解。我们提供了该方法的先验误差分析以及神经网络的稳定性分析。数值实验证实了网络预测精细有限元解的能力。我们还展示了该方法在测试域和训练域不同的问题上的泛化能力。

英文摘要:

We describe and analyze a hybrid finite element/neural network method for predicting solutions of partial differential equations. The methodology is designed for obtaining fine scale fluctuations from neural networks in a local manner. The network is capable of locally correcting a coarse finite element solution towards a fine solution taking the source term and the coarse approximation as input. Key observation is the dependency between quality of predictions and the size of training set which consists of different source terms and corresponding fine & coarse solutions. We provide the a priori error analysis of the method together with the stability analysis of the neural network. The numerical experiments confirm the capability of the network predicting fine finite element solutions. We also illustrate the generalization of the method to problems where test and training domains differ from each other.

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。

↑