Deep Adaptive Basis Galerkin Method for High-Dimensional Evolution Equations With Oscillatory Solutions

DOI10.1137/21M1468383zbMATH Open1506.65154arXiv2112.14418OpenAlexW4226337678MaRDI QIDQ5038412FDOQ5038412

Author name not available (Why is that?)

Publication date: 30 September 2022

Published in: SIAM Journal on Scientific Computing (Search for Journal in Brave)

Abstract: In this paper, we study deep neural networks (DNNs) for solving high-dimensional evolution equations with oscillatory solutions. Different from deep least-squares methods that deal with time and space variables simultaneously, we propose a deep adaptive basis Galerkin (DABG) method, which employs the spectral-Galerkin method for the time variable of oscillatory solutions and the deep neural network method for high-dimensional space variables. The proposed method can lead to a linear system of differential equations having unknown DNNs that can be trained via the loss function. We establish a posterior estimates of the solution error, which is bounded by the minimal loss function and the term $O(N^{-m})$, where $N$ is the number of basis functions and $m$ characterizes the regularity of the e'quation. We also show that if the true solution is a Barron-type function, the error bound converges to zero as $M=O(N^p)$ approaches to infinity, where $M$ is the width of the used networks, and $p$ is a positive constant. Numerical examples, including high-dimensional linear evolution equations and the nonlinear Allen-Cahn equation, are presented to demonstrate the performance of the proposed DABG method is better than that of existing DNNs.

Full work available at URL: https://arxiv.org/abs/2112.14418

zbMATH Keywords

Galerkin methodmethod of linesparabolic equationhyperbolic equationdeep learning in spaceLegendre polynomials in time

Mathematics Subject Classification ID

Artificial neural networks and deep learning (68T07) Oscillation, zeros of solutions, mean value theorems, etc. in context of PDEs (35B05) Fourier series in special orthogonal functions (Legendre polynomials, Walsh functions, etc.) (42C10) Error bounds for initial value and initial-boundary value problems involving PDEs (65M15) Method of lines for initial value and initial-boundary value problems involving PDEs (65M20) Finite element, Rayleigh-Ritz and Galerkin methods for initial value and initial-boundary value problems involving PDEs (65M60) Rate of convergence, degree of approximation (41A25) Spectral, collocation and related methods for initial value and initial-boundary value problems involving PDEs (65M70)

Cites Work

• Title not available (Why is that?)
• Title not available (Why is that?)
• DGM: a deep learning algorithm for solving partial differential equations
• Universal approximation bounds for superpositions of a sigmoidal function
• Theory and practice of finite elements.
• Spectral Methods
• Elliptic Partial Differential Equations of Second Order
• Approximation rates for neural networks with general activation functions
• Solving parametric PDE problems with artificial neural networks
• Solving high-dimensional partial differential equations using deep learning
• Deep learning-based numerical methods for high-dimensional parabolic partial differential equations and backward stochastic differential equations
• The Deep Ritz Method: a deep learning-based numerical algorithm for solving variational problems
• Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
• Weak adversarial networks for high-dimensional partial differential equations
• Deep Network With Approximation Error Being Reciprocal of Width to Power of Square Root of Depth
• Approximation by Combinations of ReLU and Squared ReLU Ridge Functions With <inline-formula> <tex-math notation="LaTeX">$\ell^1$ </tex-math> </inline-formula> and <inline-formula> <tex-math notation="LaTeX">$\ell^0$ </tex-math> </inline-formula> Controls
• Neural network approximation: three hidden layers are enough
• Deep Splitting Method for Parabolic PDEs
• The Barron space and the flow-induced function spaces for neural network models
• Structure probing neural network deflation

Uses Software

• OPQ
• DGM

Recommendations

• A new class of uniformly accurate numerical schemes for highly oscillatory evolution equations 👍 👎
• High-Order Adaptive Galerkin Methods 👍 👎
• Adaptive Galerkin finite element methods for the wave equation 👍 👎
• Multi-Adaptive Galerkin Methods for ODEs I 👍 👎
• Multi-Adaptive Galerkin Methods for ODEs II: implementation and Applications 👍 👎
• Adaptive high-order splitting methods for systems of nonlinear evolution equations with periodic boundary conditions 👍 👎
• Local adaptive Galerkin bases for large-dimensional dynamical systems 👍 👎
• Adaptive high-order finite-difference method for nonlinear wave problems 👍 👎
• Adaptive deep neural networks methods for high-dimensional partial differential equations 👍 👎
• Extensions of the deep Galerkin method 👍 👎

This page was built for publication: Deep Adaptive Basis Galerkin Method for High-Dimensional Evolution Equations With Oscillatory Solutions

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5038412)