A multi-stage deep learning based algorithm for multiscale modelreduction

A multi-stage deep learning based algorithm for multiscale modelreduction
复制标题

一种基于多阶段深度学习的多尺度模型缩减算法

DOI:
10.1016/j.cam.2021.113506
复制
发表时间:
2020
期刊:
J. Comput. Appl. Math.
影响因子:
--
通讯作者:
Zecheng Zhang
Zecheng Zhang
中科院分区:
--
文献类型:
--
作者:
Eric T. Chung;W. Leung;Sai;Zecheng Zhang

文献摘要

参考文献

被引文献

相似文献

在这项工作中,我们提出了一种多阶段训练策略,用于开发应用于多尺度特征问题的深度学习算法。所提出的策略的每个阶段共享(几乎)相同的网络结构,并预测相同的降阶模型的多尺度问题。前一级的输出将与当前级的中间层相结合。我们用数值方法证明了使用不同的降阶模型作为每个阶段的输入可以改善训练,并且我们提出了几种向系统中添加不同信息的方法。这些方法包括数学多尺度模型简化和网络方法,但我们发现,数学方法是一个系统的解耦信息的方法,并给出了最好的结果。最后,我们验证了我们的训练方法上的时间依赖的非线性问题和稳态模型。
In this work, we propose a multi-stage training strategy for the development of deep learning algorithms applied to problems with multiscale features. Each stage of the proposed strategy shares an (almost) identical network structure and predicts the same reduced order model of the multiscale problem. The output of the previous stage will be combined with an intermediate layer for the current stage. We numerically show that using different reduced order models as inputs of each stage can improve the training and we propose several ways of adding different information into the systems. These methods include mathematical multiscale model reductions and network approaches; but we found that the mathematical approach is a systematical way of decoupling information and gives the best result. We finally verified our training methodology on a time dependent nonlinear problem and a steady state model.
DOI: 10.3390/math8050720
发表时间: 2020-04
期刊: ArXiv
影响因子: --
作者:
Zecheng Zhang;Eric T. Chung;Y. Efendiev;W. Leung
通讯作者: Zecheng Zhang;Eric T. Chung;Y. Efendiev;W. Leung
使用非局部多连续放大技术学习非线性多尺度模拟中的宏观参数
DOI: 10.1016/j.jcp.2020.109323
发表时间: 2020
影响因子: 4.1
作者:
Vasilyeva, Maria;Leung, Wing T.;Chung, Eric T.;Efendiev, Yalchin;Wheeler, Mary
通讯作者: Wheeler, Mary