Computing Multi-Eigenpairs of High-Dimensional Eigenvalue Problems Using Tensor Neural Networks

DOI10.1016/J.JCP.2024.112928MaRDI QIDQ6437422zbMATH OpenFDO

Authors Yifan Wang, Hehu Xie

Publication date 21 May 2023

Full work available at URL https://doi.org/10.1016/J.JCP.2024.112928

Mathematics Subject Classification ID

Numerical linear algebra (65Fxx) Numerical methods for partial differential equations, initial value and time-dependent initial-boundary value problems (65Mxx) Artificial intelligence (68Txx)

Abstract: In this paper, we propose a type of tensor-neural-network-based machine learning method to compute multi-eigenpairs of high dimensional eigenvalue problems without Monte-Carlo procedure. Solving multi-eigenvalues and their corresponding eigenfunctions is one of the basic tasks in mathematical and computational physics. With the help of tensor neural network and deep Ritz method, the high dimensional integrations included in the loss functions of the machine learning process can be computed with high accuracy. The high accuracy of high dimensional integrations can improve the accuracy of the machine learning method for computing multi-eigenpairs of high dimensional eigenvalue problems. Here, we introduce the tensor neural network and design the machine learning method for computing multi-eigenpairs of the high dimensional eigenvalue problems. The proposed numerical method is validated with plenty of numerical examples.

Recommendations

• Solving high-dimensional eigenvalue problems using deep neural networks: a diffusion Monte Carlo like approach
• Low-rank Riemannian eigensolver for high-dimensional Hamiltonians
• Solving high-dimensional partial differential equations using tensor neural network and a posteriori error estimators
• Subspace methods with local refinements for eigenvalue computation using low-rank tensor-train format
• Preconditioned low-rank methods for high-dimensional elliptic PDE eigenvalue problems

Cited in

(16)

• Solving high-dimensional eigenvalue problems using deep neural networks: a diffusion Monte Carlo like approach
• Solving eigenvalue PDEs of metastable diffusion processes using artificial neural networks
• Solving high-dimensional partial differential equations using tensor neural network and a posteriori error estimators
• Generalization error estimates of a machine learning method for solving high-dimensional Schrödinger eigenvalue problems
• FieldTNN-based machine learning method for Maxwell eigenvalue problems
• Solving time-fractional partial integro-differential equations using tensor neural network
• A radial basis function neural network approach for solving a diffusion partial differential equation efficiently
• Tensor neural network-based machine learning method for elliptic multiscale problems
• Tensor neural network combined with Nitsche method for solving interface problems
• A structure-preserving PINN with embedded periodic boundary layer and adaptively enforced initial conditions for geometric flows
• A positivity-preserving subspace method based on neural networks for solving diffusion equations in the weak form
• Hierarchical rank-evolving representation for physics-informed neural networks
• Domain decomposition subspace neural network method for solving linear and nonlinear partial differential equations
• Subspace method based on neural networks for solving eigenvalue problems
• A comparative study of sampling techniques for generating correlated multivariate normal random variates
• Solving singularly perturbed eigenvalue problems via TFPM-inspired neural networks

This page was built for publication: Computing Multi-Eigenpairs of High-Dimensional Eigenvalue Problems Using Tensor Neural Networks

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6437422)