A decision-making machine learning approach in Hermite spectral approximations of partial differential equations

04/05/2021
by   Lorella Fatone, et al.
0

The accuracy and effectiveness of Hermite spectral methods for the numerical discretization of partial differential equations on unbounded domains, are strongly affected by the amplitude of the Gaussian weight function employed to describe the approximation space. This is particularly true if the problem is under-resolved, i.e., there are no enough degrees of freedom. The issue becomes even more crucial when the equation under study is time-dependent, forcing in this way the choice of Hermite functions where the corresponding weight depends on time. In order to adapt dynamically the approximation space, it is here proposed an automatic decision-making process that relies on machine learning techniques, such as deep neural networks and support vector machines. The algorithm is numerically tested with success on a simple 1D problem, but the main goal is its exportability in the context of more serious applications.

READ FULL TEXT
POST COMMENT

Comments

There are no comments yet.

Authors

page 1

page 2

page 3

page 4

07/08/2019

Physics Informed Extreme Learning Machine (PIELM) -- A rapid method for the numerical solution of partial differential equations

There has been rapid progress recently on the application of deep networ...
08/30/2019

Matrix equation techniques for certain evolutionary partial differential equations

We show that the discrete operator stemming from the time and space disc...
03/29/2017

Numerical Gaussian Processes for Time-dependent and Non-linear Partial Differential Equations

We introduce the concept of numerical Gaussian processes, which we defin...
08/02/2017

Hidden Physics Models: Machine Learning of Nonlinear Partial Differential Equations

While there is currently a lot of enthusiasm about "big data", useful da...
11/09/2021

Machine-learning custom-made basis functions for partial differential equations

Spectral methods are an important part of scientific computing's arsenal...
06/28/2020

Monolithic Multigrid for Magnetohydrodynamics

The magnetohydrodynamics (MHD) equations model a wide range of plasma ph...
04/06/2020

Massively Parallel Stencil Strategies for Radiation Transport Moment Model Simulations

The radiation transport equation is a mesoscopic equation in high dimens...
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.