Saved in:
Bibliographic Details
Main Authors: Mojarrad, Fatemeh Nassajian, Veiga, Maria Han, Hesthaven, Jan S., Öffner, Philipp
Format: Preprint
Published: 2022
Subjects:
Online Access:https://arxiv.org/abs/2210.16945
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917703922483200
author Mojarrad, Fatemeh Nassajian
Veiga, Maria Han
Hesthaven, Jan S.
Öffner, Philipp
author_facet Mojarrad, Fatemeh Nassajian
Veiga, Maria Han
Hesthaven, Jan S.
Öffner, Philipp
contents The choice of the shape parameter highly effects the behaviour of radial basis function (RBF) approximations, as it needs to be selected to balance between ill-condition of the interpolation matrix and high accuracy. In this paper, we demonstrate how to use neural networks to determine the shape parameters in RBFs. In particular, we construct a multilayer perceptron trained using an unsupervised learning strategy, and use it to predict shape parameters for inverse multiquadric and Gaussian kernels. We test the neural network approach in RBF interpolation tasks and in a RBF-finite difference method in one and two-space dimensions, demonstrating promising results.
format Preprint
id arxiv_https___arxiv_org_abs_2210_16945
institution arXiv
publishDate 2022
record_format arxiv
spellingShingle A new variable shape parameter strategy for RBF approximation using neural networks
Mojarrad, Fatemeh Nassajian
Veiga, Maria Han
Hesthaven, Jan S.
Öffner, Philipp
Numerical Analysis
The choice of the shape parameter highly effects the behaviour of radial basis function (RBF) approximations, as it needs to be selected to balance between ill-condition of the interpolation matrix and high accuracy. In this paper, we demonstrate how to use neural networks to determine the shape parameters in RBFs. In particular, we construct a multilayer perceptron trained using an unsupervised learning strategy, and use it to predict shape parameters for inverse multiquadric and Gaussian kernels. We test the neural network approach in RBF interpolation tasks and in a RBF-finite difference method in one and two-space dimensions, demonstrating promising results.
title A new variable shape parameter strategy for RBF approximation using neural networks
topic Numerical Analysis
url https://arxiv.org/abs/2210.16945