Solving Elliptic Problems with Singular Sources Using Singularity Splitting Deep Ritz Method

DOI10.1137/22m1520840zbMath1528.65110arXiv2209.02931OpenAlexW4385659965MaRDI QIDQ6095431

Unnamed Author, Zhi Zhou, Bangti Jin

Publication date: 8 September 2023

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

Full work available at URL: https://arxiv.org/abs/2209.02931

zbMATH Keywords

neural networkspenalty methodsingular sourcevariable coefficient Poisson equationdeep Ritz method

Mathematics Subject Classification ID

Artificial neural networks and deep learning (68T07) Stability and convergence of numerical methods for boundary value problems involving PDEs (65N12) Finite element, Rayleigh-Ritz and Galerkin methods for boundary value problems involving PDEs (65N30) Laplace operator, Helmholtz equation (reduced wave equation), Poisson equation (35J05)

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Discretization of Dirac delta functions in level set methods
• Weak adversarial networks for high-dimensional partial differential equations
• A primal dual active set with continuation algorithm for the \(\ell^0\)-regularized optimization problem
• The Green function for uniformly elliptic equations
• The Deep Ritz Method: a deep learning-based numerical algorithm for solving variational problems
• Theory and practice of finite elements.
• DGM: a deep learning algorithm for solving partial differential equations
• \textit{hp}-VPINNs: variational physics-informed neural networks with domain decomposition
• Approximation rates for neural networks with encodable weights in smoothness spaces
• A mixed formulation for physics-informed neural networks as a potential solver for engineering problems in heterogeneous domains: comparison with finite element method
• Uniform convergence guarantees for the deep Ritz method for nonlinear problems
• Self-adaptive physics-informed neural networks
• Least-squares ReLU neural network (LSNN) method for linear advection-reaction equation
• When and why PINNs fail to train: a neural tangent kernel perspective
• Scientific machine learning through physics-informed neural networks: where we are and what's next
• On the use of singular functions with finite element approximations
• Error bounds for approximations with deep ReLU networks
• Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
• On regularizations of the Dirac delta distribution
• Error-bounds for finite element method
• Finite element convergence for singular data
• A comprehensive study of non-adaptive and residual-based adaptive sampling for physics-informed neural networks
• DAS-PINNs: a deep adaptive sampling method for solving high-dimensional partial differential equations
• A Finite Element Method Using Singular Functions for the Poisson Equation: Corner Singularities
• Finite Element Approximation of Elliptic Problems with Dirac Measure Terms in Weighted Spaces: Applications to One- and Three-dimensional Coupled Problems
• Optimal A Priori Error Estimates for an Elliptic Problem with Dirac Right-Hand Side
• The extended/generalized finite element method: An overview of the method and its applications
• A Priori Mesh Grading for an Elliptic Problem with Dirac Right-Hand Side
• ON THE COUPLING OF 1D AND 3D DIFFUSION-REACTION EQUATIONS: APPLICATION TO TISSUE PERFUSION PROBLEMS
• Electrical impedance tomography and Calderón's problem
• Inverse boundary value problems at the boundary—continuous dependence
• A Singularly mixed boundary value problem
• Introduction to Numerical Continuation Methods
• Optimization Methods for Large-Scale Machine Learning
• Robin Approximation of Dirichlet Boundary Value Problems
• A Limited Memory Algorithm for Bound Constrained Optimization
• Properties of Discrete Delta Functions and Local Convergence of the Immersed Boundary Method
• Neural Network Learning
• Understanding and Mitigating Gradient Flow Pathologies in Physics-Informed Neural Networks
• Splitting method for elliptic equations with line sources
• A Mixed Approach to the Poisson Problem with Line Sources
• The Gap between Theory and Practice in Function Approximation with Deep Neural Networks
• Algorithms for solving high dimensional PDEs: from nonlinear Monte Carlo to machine learning
• Convergence Rate Analysis for Deep Ritz Method
• Imaging conductivity from current density magnitude using neural networks*
• Understanding Machine Learning
• Neural network approximation
• Finite basis physics-informed neural networks (FBPINNs): a scalable domain decomposition approach for solving differential equations

This page was built for publication: Solving Elliptic Problems with Singular Sources Using Singularity Splitting Deep Ritz Method