Numerical solution of the parametric diffusion equation by deep neural networks
From MaRDI portal
Elliptic equations and elliptic systems (35J99) Rate of convergence, degree of approximation (41A25) Approximation by other special function classes (41A30) Finite element, Rayleigh-Ritz and Galerkin methods for boundary value problems involving PDEs (65N30) Learning and adaptive systems in artificial intelligence (68T05)
Abstract: We perform a comprehensive numerical study of the effect of approximation-theoretical results for neural networks on practical learning problems in the context of numerical analysis. As the underlying model, we study the machine-learning-based solution of parametric partial differential equations. Here, approximation theory predicts that the performance of the model should depend only very mildly on the dimension of the parameter space and is determined by the intrinsic dimension of the solution manifold of the parametric partial differential equation. We use various methods to establish comparability between test-cases by minimizing the effect of the choice of test-cases on the optimization and sampling aspects of the learning problem. We find strong support for the hypothesis that approximation-theoretical effects heavily influence the practical behavior of learning problems in numerical analysis.
Recommendations
- Solving parametric PDE problems with artificial neural networks
- Solving parametric partial differential equations with deep rectified quadratic unit neural networks
- Efficient approximation of solutions of parametric linear transport equations by ReLU DNNs
- Model reduction and neural networks for parametric PDEs
- A finite element based deep learning solver for parametric PDEs
Cites work
- A mixed ℓ1 regularization approach for sparse simultaneous approximation of parameterized PDEs
- A proof that rectified deep neural networks overcome the curse of dimensionality in the numerical approximation of semilinear heat equations
- An artificial neural network framework for reduced order modeling of transient flows
- An energy approach to the solution of partial differential equations in computational mechanics via machine learning: concepts, implementation and applications
- An introduction to computational stochastic PDEs
- Analytic regularity and polynomial approximation of stochastic, parametric elliptic multiscale PDEs
- Approximation of high-dimensional parametric PDEs
- Approximation of infinitely differentiable multivariate functions is intractable
- Certified reduced basis methods for parametrized partial differential equations
- Compressive sensing Petrov-Galerkin approximation of high-dimensional parametric operator equations
- Data driven approximation of parametrized PDEs by reduced basis and neural networks
- Data-driven discovery of PDEs in complex datasets
- Deep hidden physics models: deep learning of nonlinear partial differential equations
- Deep learning in high dimension: neural network expression rates for generalized polynomial chaos expansions in UQ
- Deep learning-based numerical methods for high-dimensional parabolic partial differential equations and backward stochastic differential equations
- Deep ReLU networks and high-order finite element methods
- Deep UQ: learning deep neural network surrogate models for high dimensional uncertainty quantification
- DeepXDE: a deep learning library for solving differential equations
- DGM: a deep learning algorithm for solving partial differential equations
- Efficient approximation of solutions of parametric linear transport equations by ReLU DNNs
- Equivalence of approximation by convolutional neural networks and fully-connected networks
- Error bounds for approximations with deep ReLU networks
- scientific article; zbMATH DE number 43550 (Why is no real title available?)
- Inferring solutions of differential equations using noisy multi-fidelity data
- Kolmogorov widths and low-rank approximations of parametric elliptic PDEs
- Learning Theory
- Machine learning approximation algorithms for high-dimensional fully nonlinear partial differential equations and second-order backward stochastic differential equations
- Machine learning for fast and reliable solution of time-dependent differential equations
- Machine learning of linear differential equations using Gaussian processes
- Model reduction of dynamical systems on nonlinear manifolds using deep convolutional autoencoders
- Non-intrusive reduced order modeling of nonlinear problems using neural networks
- On the mathematical foundations of learning
- On the Theory of Dynamic Programming
- Optimal approximation of piecewise smooth functions using deep ReLU neural networks
- Optimal approximation with sparsely connected deep neural networks
- Parametric PDEs: sparse or low-rank approximations?
- Provable approximation properties for deep neural networks
- Reduced basis approximation and a posteriori error estimation for affinely parametrized elliptic coercive partial differential equations. Application to transport and continuum mechanics.
- Reduced basis methods for partial differential equations. An introduction
- Solving high-dimensional partial differential equations using deep learning
- The deep Ritz method: a deep learning-based numerical algorithm for solving variational problems
- Theory of deep convolutional neural networks: downsampling
- Universal approximation bounds for superpositions of a sigmoidal function
- Variational Monte Carlo -- bridging concepts of machine learning and high-dimensional partial differential equations
Cited in
(58)- Efficient approximation of solutions of parametric linear transport equations by ReLU DNNs
- Model reduction and neural networks for parametric PDEs
- CAS4DL: Christoffel adaptive sampling for function approximation via deep learning
- Solving parametric partial differential equations with deep rectified quadratic unit neural networks
- A theoretical analysis of deep neural networks and parametric PDEs
- Meta-learning pseudo-differential operators with deep neural networks
- Self-adaptive deep neural network: numerical approximation to functions and PDEs
- Solving multiscale steady radiative transfer equation using neural networks with uniform stability
- Deep neural networks based temporal-difference methods for high-dimensional parabolic partial differential equations
- Deep neural network approach to forward-inverse problems
- Neural-net-induced Gaussian process regression for function approximation and PDE solution
- ConvPDE-UQ: convolutional neural networks with quantified uncertainty for heterogeneous elliptic partial differential equations on varied domains
- Physics informed by deep learning: numerical solutions of modified Korteweg-de Vries equation
- Variational Monte Carlo -- bridging concepts of machine learning and high-dimensional partial differential equations
- A finite element based deep learning solver for parametric PDEs
- Numerical wave propagation aided by deep learning
- Long-time integration of parametric evolution equations with physics-informed DeepONets
- Computation and learning in high dimensions. Abstracts from the workshop held August 1--7, 2021 (hybrid meeting)
- An overview on deep learning-based approximation methods for partial differential equations
- The Random Feature Model for Input-Output Maps between Banach Spaces
- Emulated digital CNN-UM solution of partial differential equations
- Automated design parameter selection for neural networks solving coupled partial differential equations with discontinuities
- A QMC-Deep Learning Method for Diffusivity Estimation in Random Domains
- Solving parametric PDE problems with artificial neural networks
- Convergence bounds for empirical nonlinear least-squares
- A deep learning approach to Reduced Order Modelling of parameter dependent partial differential equations
- Deep neural network surrogates for nonsmooth quantities of interest in shape uncertainty quantification
- Simultaneous neural network approximation for smooth functions
- Numerical solution of nonlinear stochastic differential equations with fractional Brownian motion using fractional-order Genocchi deep neural networks
- Deep ReLU neural network approximation in Bochner spaces and applications to parametric PDEs
- An introduction to the mathematics of deep learning
- Deep learning methods for partial differential equations and related parameter identification problems
- Mesh-informed neural networks for operator learning in finite element spaces
- A New Certified Hierarchical and Adaptive RB-ML-ROM Surrogate Model for Parametrized PDEs
- Prediction of numerical homogenization using deep learning for the Richards equation
- Exponential Convergence of Deep Operator Networks for Elliptic Partial Differential Equations
- Multigrid-Augmented Deep Learning Preconditioners for the Helmholtz Equation
- Collocation approximation by deep neural ReLU networks for parametric and stochastic PDEs with lognormal inputs
- Model reduction of coupled systems based on non-intrusive approximations of the boundary response maps
- The mathematics of artificial intelligence
- Error assessment of an adaptive finite elements -- neural networks method for an elliptic parametric PDE
- Convergence rate of DeepONets for learning operators arising from advection-diffusion equations
- Simultaneous approximation of a smooth function and its derivatives by deep neural networks with piecewise-polynomial activations
- Learning quantities of interest from parametric PDEs: an efficient neural-weighted minimal residual approach
- Operator learning using random features: a tool for scientific computing
- One-shot learning of surrogates in PDE-constrained optimization under uncertainty
- Error analysis for deep neural network approximations of parametric hyperbolic conservation laws
- Numerical solution of Poisson partial differential equation in high dimension using two-layer neural networks
- Comparing reduced order surrogate models for electrical impedance tomography
- A low-rank neural representation of entropy solutions
- Numerical solution of mixed-dimensional PDEs using a neural preconditioner
- Operator theory, kernels, and feedforward neural networks
- A least-squares-based neural network (LS-Net) for solving linear parametric PDEs
- Application of dimension truncation error analysis to high-dimensional function approximation in uncertainty quantification
- Accelerating the convergence of Newton's method for nonlinear elliptic PDEs using Fourier neural operators
- The troublesome kernel: on hallucinations, no free lunches, and the accuracy-stability tradeoff in inverse problems
- Multilayer perceptrons as function approximators for analytical solutions of the diffusion equation
- Translating numerical concepts for PDEs into neural architectures
This page was built for publication: Numerical solution of the parametric diffusion equation by deep neural networks
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2049099)
