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.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is the efficient and accurate computation of multiple eigenpairs (i.e., multiple eigenvalues and their corresponding eigenfunctions) in high - dimensional eigenvalue problems. Specifically, the paper proposes a machine - learning method based on Tensor Neural Network (TNN) for computing multiple eigenpairs of high - dimensional eigenvalue problems without using the Monte Carlo process. Traditional numerical methods such as the finite - difference method, the finite - element method and the spectral method perform well when dealing with low - dimensional Schrödinger equations, but when faced with high - dimensional problems, these methods will be affected by the curse of dimensionality, that is, as the dimension increases, the degrees of freedom and the computational complexity increase exponentially. Although the Monte Carlo method can handle high - dimensional problems, its convergence speed is low, resulting in difficulty in obtaining high - precision and stable solutions.
The main contributions of the paper are as follows:
1. **Proposing a machine - learning method based on TNN**: This method utilizes the tensor neural network and the deep Ritz method, and can efficiently and accurately compute multiple eigenpairs of high - dimensional eigenvalue problems.
2. **Avoiding the inefficiency of the Monte Carlo method**: Through TNN, high - dimensional integrals can be decomposed into one - dimensional integrals, and can be computed by a high - precision Gaussian - type quadrature scheme, thus overcoming the curse of dimensionality.
3. **Designing a loss function suitable for the computation of multiple eigenpairs**: Different from the traditional method based on the Rayleigh quotient, the loss function of this method is more similar to the classical method, and can better ensure the orthogonality and normalization conditions of the solution.
The paper verifies the effectiveness and accuracy of the proposed method through a large number of numerical experiments, and shows the advantages of TNN in solving high - dimensional eigenvalue problems.