A unified deep artificial neural network approach to partial differential equations in complex geometries

Abstract

In this paper we use deep feedforward artificial neural networks to approximate solutions to partial differential equations in complex geometries. We show how to modify the backpropagation algorithm to compute the partial derivatives of the network output with respect to the space variables which is needed to approximate the differential operator. The method is based on an ansatz for the solution which requires nothing but feedforward neural networks and an unconstrained gradient based optimization method such as gradient descent or a quasi-Newton method. We show an example where classical mesh based methods cannot be used and neural networks can be seen as an attractive alternative. Finally, we highlight the benefits of deep compared to shallow neural networks and device some other convergence enhancing techniques.

arXiv comment35 pages, 12 figures

References in corpus (6)

• Artificial Neural Networks for Solving Ordinary and Partial Differential Equations
• DGM: A deep learning algorithm for solving partial differential equations
• 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
• Proceedings of the Sixteenth Conference on Uncertainty in Artificial Intelligence (2000)
• Computational Investigation of Low-Discrepancy Sequences in Simulation Algorithms for Bayesian Networks

Cited by in corpus (153)

• DeepXDE: A deep learning library for solving differential equations
• An Energy Approach to the Solution of Partial Differential Equations in Computational Mechanics via Machine Learning: Concepts, Implementation and Applications
• Surrogate Modeling for Fluid Flows Based on Physics-Constrained Deep Learning Without Simulation Data
• Physics-Constrained Deep Learning for High-dimensional Surrogate Modeling and Uncertainty Quantification without Labeled Data
• hp-VPINNs: Variational Physics-Informed Neural Networks With Domain Decomposition
• PhyGeoNet: Physics-Informed Geometry-Adaptive Convolutional Neural Networks for Solving Parameterized Steady-State PDEs on Irregular Domain
• A Deep Collocation Method for the Bending Analysis of Kirchhoff Plate
• Weak Adversarial Networks for High-dimensional Partial Differential Equations
• SciANN: A Keras/Tensorflow wrapper for scientific computations and physics-informed deep learning using artificial neural networks
• Physics-Informed Neural Networks for Multiphysics Data Assimilation with Application to Subsurface Transport
• CAN-PINN: A Fast Physics-Informed Neural Network Based on Coupled-Automatic-Numerical Differentiation Method
• Finite Basis Physics-Informed Neural Networks (FBPINNs): a scalable domain decomposition approach for solving differential equations
• Modeling the Dynamics of PDE Systems with Physics-Constrained Deep Auto-Regressive Networks
• Efficient training of physics-informed neural networks via importance sampling
• Physics-Informed Neural Network for Modelling the Thermochemical Curing Process of Composite-Tool Systems During Manufacture
• Physics-informed graph neural Galerkin networks: A unified framework for solving PDE-governed forward and inverse problems
• Physics informed neural networks for continuum micromechanics
• Solving Many-Electron Schrödinger Equation Using Deep Neural Networks
• A fast and accurate physics-informed neural network reduced order model with shallow masked autoencoder
• Biologically-informed neural networks guide mechanistic modeling from sparse experimental data
• A Method for Representing Periodic Functions and Enforcing Exactly Periodic Boundary Conditions with Deep Neural Networks
• Deep-learning-based surrogate flow modeling and geological parameterization for data assimilation in 3D subsurface flow
• Simulator-free Solution of High-Dimensional Stochastic Elliptic Partial Differential Equations using Deep Neural Networks
• Deep Learning in Deterministic Computational Mechanics
• Algorithms for Solving High Dimensional PDEs: From Nonlinear Monte Carlo to Machine Learning
• Solving Fokker-Planck equation using deep learning
• PFNN: A Penalty-Free Neural Network Method for Solving a Class of Second-Order Boundary-Value Problems on Complex Geometries
• Physics-Informed Neural Network Method for Forward and Backward Advection-Dispersion Equations
• The Neural Particle Method -- An Updated Lagrangian Physics Informed Neural Network for Computational Fluid Dynamics
• Deep Nitsche Method: Deep Ritz Method with Essential Boundary Conditions
• Deep splitting method for parabolic PDEs
• Distributed physics informed neural network for data-efficient solution to partial differential equations
• Solving Electrical Impedance Tomography with Deep Learning
• Deep least-squares methods: an unsupervised learning-based numerical method for solving elliptic PDEs
• SPINN: Sparse, Physics-based, and partially Interpretable Neural Networks for PDEs
• Learning in Sinusoidal Spaces with Physics-Informed Neural Networks
• CENN: Conservative energy method based on neural networks with subdomains for solving variational problems involving heterogeneous and complex geometries
• Physics-Driven Regularization of Deep Neural Networks for Enhanced Engineering Design and Analysis
• An overview on deep learning-based approximation methods for partial differential equations
• Enabling real-time multi-messenger astrophysics discoveries with deep learning
• On some neural network architectures that can represent viscosity solutions of certain high dimensional Hamilton--Jacobi partial differential equations
• Evolutional Deep Neural Network
• SelectNet: Self-paced Learning for High-dimensional Partial Differential Equations
• Int-Deep: A Deep Learning Initialized Iterative Method for Nonlinear Problems
• BINN: A deep learning approach for computational mechanics problems based on boundary integral equations
• Physics Informed Neural Fields for Smoke Reconstruction with Sparse Data
• HiDeNN-PGD: reduced-order hierarchical deep learning neural networks
• $\Delta$-PINNs: physics-informed neural networks on complex geometries
• Trainability and Accuracy of Neural Networks: An Interacting Particle System Approach
• Mosaic Flows: A Transferable Deep Learning Framework for Solving PDEs on Unseen Domains
• BCR-Net: a neural network based on the nonstandard wavelet form
• Multi-Output Physics-Informed Neural Networks for Forward and Inverse PDE Problems with Uncertainties
• Two-Layer Neural Networks for Partial Differential Equations: Optimization and Generalization Theory
• Learning nonlocal constitutive models with neural networks
• Overcoming the curse of dimensionality in the approximative pricing of financial derivatives with default risks
• DCEM: A deep complementary energy method for solid mechanics
• Overcoming the curse of dimensionality in the numerical approximation of Allen-Cahn partial differential equations via truncated full-history recursive multilevel Picard approximations
• Neural Networks Trained to Solve Differential Equations Learn General Representations
• Neural Galerkin Schemes with Active Learning for High-Dimensional Evolution Equations
• Hyperbolic compactification of M-theory and de Sitter quantum gravity
• On quadrature rules for solving Partial Differential Equations using Neural Networks
• Reduced and All-at-Once Approaches for Model Calibration and Discovery in Computational Solid Mechanics
• Trend to Equilibrium for the Kinetic Fokker-Planck Equation via the Neural Network Approach
• An unsupervised latent/output physics-informed convolutional-LSTM network for solving partial differential equations using peridynamic differential operator
• Variational training of neural network approximations of solution maps for physical models
• Using neural networks to accelerate the solution of the Boltzmann equation
• GINNs: Graph-Informed Neural Networks for Multiscale Physics
• An augmented Lagrangian deep learning method for variational problems with essential boundary conditions
• Numerical simulations for full history recursive multilevel Picard approximations for systems of high-dimensional partial differential equations
• WarpPINN: Cine-MR image registration with physics-informed neural networks
• Error estimates of residual minimization using neural networks for linear PDEs
• Meta-learning Pseudo-differential Operators with Deep Neural Networks
• Uniform error estimates for artificial neural network approximations for heat equations
• Physics-informed neural wavefields with Gabor basis functions
• Physics-aware deep neural networks for surrogate modeling of turbulent natural convection
• Can Transfer Neuroevolution Tractably Solve Your Differential Equations?
• Data-Driven Finite Elements Methods: Machine Learning Acceleration of Goal-Oriented Computations
• Neural Networks Enforcing Physical Symmetries in Nonlinear Dynamical Lattices: The Case Example of the Ablowitz-Ladik Model
• Exact imposition of boundary conditions with distance functions in physics-informed deep neural networks
• Deep Learning for Cardiologist-level Myocardial Infarction Detection in Electrocardiograms
• Overcoming the curse of dimensionality in the numerical approximation of parabolic partial differential equations with gradient-dependent nonlinearities
• Physics-Guided Neural Networks for Intraventricular Vector Flow Mapping
• A multiscale neural network based on hierarchical matrices
• Neural network architectures using min-plus algebra for solving certain high dimensional optimal control problems and Hamilton-Jacobi PDEs
• A priori and a posteriori error estimates for the Deep Ritz method applied to the Laplace and Stokes problem
• Deep Ritz Method with Adaptive Quadrature for Linear Elasticity
• Enhancing approximation abilities of neural networks by training derivatives
• Structure Probing Neural Network Deflation
• Butterfly-Net: Optimal Function Representation Based on Convolutional Neural Networks
• Solving PDE-constrained Control Problems Using Operator Learning
• Deep neural network approximations for Monte Carlo algorithms
• Mutual Information for Explainable Deep Learning of Multiscale Systems
• Improved Surrogate Modeling of Fluid Dynamics with Physics-Informed Neural Networks
• Neural Networks to solve Partial Differential Equations: a Comparison with Finite Elements
• CD-ROM: Complemented Deep-Reduced Order Model
• Enforcing Deterministic Constraints on Generative Adversarial Networks for Emulating Physical Systems
• Solving for high dimensional committor functions using artificial neural networks
• Convergence rate of DeepONets for learning operators arising from advection-diffusion equations
• Physics-Informed Holomorphic Neural Networks (PIHNNs): Solving Linear Elasticity Problems
• The Deep Minimizing Movement Scheme
• Generalised multilevel Picard approximations
• Microseismic source imaging using physics-informed neural networks with hard constraints
• Neural Control of Discrete Weak Formulations: Galerkin, Least-Squares and Minimal-Residual Methods with Quasi-Optimal Weights
• Extreme learning machine collocation for the numerical solution of elliptic PDEs with sharp gradients
• A mixed formulation for physics-informed neural networks as a potential solver for engineering problems in heterogeneous domains: comparison with finite element method
• Inference of neutrino flavor evolution through data assimilation and neural differential equations
• Neural networks catching up with finite differences in solving partial differential equations in higher dimensions
• Extreme Theory of Functional Connections: A Physics-Informed Neural Network Method for Solving Parametric Differential Equations
• Self-adaptive deep neural network: Numerical approximation to functions and PDEs
• Solving Optical Tomography with Deep Learning
• Geometry-aware framework for deep energy method: an application to structural mechanics with hyperelastic materials
• Fixed-energy inverse scattering with radial basis function neural networks and its application to neutron-alpha interactions
• Cell-average based neural network method for hyperbolic and parabolic partial differential equations
• LSA-PINN: Linear Boundary Connectivity Loss for Solving PDEs on Complex Geometry
• Deep adaptive sampling for surrogate modeling without labeled data
• Solving Irregular and Data-enriched Differential Equations using Deep Neural Networks
• Discovering Physics-Informed Neural Networks Model for Solving Partial Differential Equations through Evolutionary Computation
• Overcoming the curse of dimensionality for some Hamilton--Jacobi partial differential equations via neural network architectures
• Physics-Informed Graph-Mesh Networks for PDEs: A hybrid approach for complex problems
• Multi-Constitutive Neural Network for Large Deformation Poromechanics Problem
• Solving Differential Equation with Constrained Multilayer Feedforward Network
• Solving PDEs on Unknown Manifolds with Machine Learning
• Variational PINNs with tree-based integration and boundary element data in the modeling of multi-phase architected materials
• Neural Network Solutions to Differential Equations in Non-Convex Domains: Solving the Electric Field in the Slit-Well Microfluidic Device
• Deep Neural Networks Are Effective At Learning High-Dimensional Hilbert-Valued Functions From Limited Data
• Least-Squares Neural Network (LSNN) Method For Linear Advection-Reaction Equation: Discontinuity Interface
• A Neural Solver for Variational Problems on CAD Geometries with Application to Electric Machine Simulation
• Notes on Exact Boundary Values in Residual Minimisation
• Accurately Solving Physical Systems with Graph Learning
• Runge-Kutta Physics Informed Neural Networks: Formulation and Analysis
• Galerkin Neural Networks: A Framework for Approximating Variational Equations with Error Control
• Learning Density Distribution of Reachable States for Autonomous Systems
• Numerical Differentiation-based Electrophysiology-Aware Adaptive ResNet for Inverse ECG Modeling
• Learning optimal multigrid smoothers via neural networks
• Deep Ritz-Finite Element methods: Neural Network Methods trained with Finite Elements
• Physics-Informed Neural Networks for Solving Multiscale Mode-Resolved Phonon Boltzmann Transport Equation
• Deterministic and statistical calibration of constitutive models from full-field data with parametric physics-informed neural networks
• Variational Autoencoding of PDE Inverse Problems
• Adaptive Physics-Informed Neural Networks for Markov-Chain Monte Carlo
• A multiscale neural network based on hierarchical nested bases
• A Natural Deep Ritz Method for Essential Boundary Value Problems
• Deeply Learned Spectral Total Variation Decomposition
• A physics-aware deep learning model for shear band formation around collapsing pores in shocked reactive materials
• Deep Orthogonal Decompositions for Convective Nowcasting
• Generalization Error Estimates of Machine Learning Methods for Solving High Dimensional Schrödinger Eigenvalue Problems
• Probabilistic partition of unity networks for high-dimensional regression problems
• Efficient Shallow Ritz Method For 1D Diffusion Problems
• Unfitted finite element interpolated neural networks
• Discovering Phase Field Models from Image Data with the Pseudo-spectral Physics Informed Neural Networks
• Neural Stream Functions
• Efficient Shallow Ritz Method For 1D Diffusion-Reaction Problems
• A pre-training deep learning method for simulating the large bending deformation of bilayer plates
• Length Learning for Planar Euclidean Curves