arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1807.01883math.NAcs.NA
收藏

基于递阶矩阵的多尺度神经网络

A multiscale neural network based on hierarchical matrices

Yuwei Fan, Lin Lin, Lexing Ying, Leonardo Zepeda-Nunez

摘要:

在这项工作中,我们介绍了一种新的多尺度人工神经网络的基础上的结构的$\mathcal{H}$ -矩阵。该网络通过在每个空间尺度上引入局部深度神经网络,将后者推广到非线性情况。数值结果表明,该网络能够有效地逼近离散非线性偏微分方程的离散非线性映射,如非线性薛定谔方程和Kohn-Sham密度泛函理论.

英文摘要:

In this work we introduce a new multiscale artificial neural network based on the structure of $\mathcal{H}$-matrices. This network generalizes the latter to the nonlinear case by introducing a local deep neural network at each spatial scale. Numerical results indicate that the network is able to efficiently approximate discrete nonlinear maps obtained from discretized nonlinear partial differential equations, such as those arising from nonlinear Schrödinger equations and the Kohn-Sham density functional theory.

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。

↑