Deep ReLU neural network approximation in Bochner spaces and applications to parametric PDEs

DOI10.1016/J.JCO.2023.101779MaRDI QIDQ6062166zbMATH OpenOpenAlexFDO

Authors Dinh Dũng, Van Kien Nguyen, Duong Thanh Pham

Publication date 30 November 2023

Published in Journal of Complexity (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2111.05854

zbMATH Keywords

deep ReLU neural networkshigh-dimensional approximation in Bochner spacesparametric elliptic PDEs with random inputs

Mathematics Subject Classification ID

Rate of convergence, degree of approximation (41A25) Multidimensional problems (41A63) Probabilistic models, generic numerical methods in probability and statistics (65C20) Numerical solutions to stochastic differential and integral equations (65C30) Artificial neural networks and deep learning (68T07)

Abstract: We investigate non-adaptive methods of deep ReLU neural network approximation in Bochner spaces $L_2({mathbbU}^{i}nfty,X,mu)$ of functions on ${mathbbU}^{i}nfty$ taking values in a separable Hilbert space $X$, where ${mathbbU}^{i}nfty$ is either ${mathbbR}^{i}nfty$ equipped with the standard Gaussian probability measure, or ${mathbbI}^{i}nfty:=[-1,1{]}^{i}nfty$ equipped with the Jacobi probability measure. Functions to be approximated are assumed to satisfy a certain weighted $el{l}_2$-summability of the generalized chaos polynomial expansion coefficients with respect to the measure $mu$. We prove the convergence rate of this approximation in terms of the size of approximating deep ReLU neural networks. These results then are applied to approximation of the solution to parametric elliptic PDEs with random inputs for the lognormal and affine cases.

Recommendations

• Collocation approximation by deep neural ReLU networks for parametric and stochastic PDEs with lognormal inputs
• A theoretical analysis of deep neural networks and parametric PDEs
• Constructive deep ReLU neural network approximation
• Error bounds for approximations with deep ReLU neural networks in \(W^{s , p}\) norms
• Deep ReLU neural networks in high-dimensional approximation

Cites work

• N-term Wiener chaos approximation rates for elliptic PDEs with lognormal Gaussian random inputs
• A proof that rectified deep neural networks overcome the curse of dimensionality in the numerical approximation of semilinear heat equations
• A survey of weighted polynomial approximation with exponential weights
• A theoretical analysis of deep neural networks and parametric PDEs
• Analytic regularity and polynomial approximation of parametric and stochastic elliptic PDE's
• Approximation by superpositions of a sigmoidal function
• Approximation of high-dimensional parametric PDEs
• Approximation of smoothness classes by deep rectifier networks
• Approximation spaces of deep neural networks
• Convergence of Sparse Collocation for Functions of Countably Many Gaussian Random Variables (with Application to Elliptic PDEs)
• Convergence rates of best \(N\)-term Galerkin approximations for a class of elliptic SPDEs
• Deep learning in high dimension: neural network expression rates for generalized polynomial chaos expansions in UQ
• Deep neural network approximation for high-dimensional elliptic PDEs with boundary conditions
• Deep Neural Network Approximation Theory
• Deep neural network approximations for solutions of PDEs based on Monte Carlo algorithms
• Deep neural network expression of posterior expectations in Bayesian PDE inversion
• Deep ReLU network expression rates for option prices in high-dimensional, exponential Lévy models
• Deep ReLU neural networks in high-dimensional approximation
• Deep UQ: learning deep neural network surrogate models for high dimensional uncertainty quantification
• DGM: a deep learning algorithm for solving partial differential equations
• DNN expression rate analysis of high-dimensional PDEs: application to option pricing
• Erratum to: “Sparse-grid polynomial interpolation approximation and integration for parametric and stochastic elliptic PDEs with lognormal inputs”
• Error bounds for approximations with deep ReLU networks
• Error bounds for approximations with deep ReLU neural networks in \(W^{s , p}\) norms
• Exponential convergence of the deep neural network approximation for analytic functions
• Exponential ReLU DNN expression of holomorphic maps in high dimension
• Fully discrete approximation of parametric and stochastic elliptic PDEs
• scientific article; zbMATH DE number 17222 (Why is no real title available?)
• scientific article; zbMATH DE number 3222151 (Why is no real title available?)
• Multilayer feedforward networks are universal approximators
• Multilevel approximation of parametric and stochastic PDES
• Neural network approximation
• New error bounds for deep ReLU networks using sparse grids
• Nonlinear approximation and (deep) ReLU networks
• Numerical solution of the parametric diffusion equation by deep neural networks
• Optimal approximation of piecewise smooth functions using deep ReLU neural networks
• Proof that deep artificial neural networks overcome the curse of dimensionality in the numerical approximation of Kolmogorov partial differential equations with constant diffusion and nonlinear drift coefficients
• Sparse polynomial approximation of parametric elliptic PDEs. Part I: affine coefficients
• Sparse polynomial approximation of parametric elliptic PDEs. II: Lognormal coefficients.
• Sparse-grid polynomial interpolation approximation and integration for parametric and stochastic elliptic PDEs with lognormal inputs

Cited in

(9)

• Derivative-informed projected neural networks for high-dimensional parametric maps governed by PDEs
• A theoretical analysis of deep neural networks and parametric PDEs
• Exponential ReLU DNN expression of holomorphic maps in high dimension
• Space-time error estimates for deep neural network approximations for differential equations
• Approximations with deep neural networks in Sobolev time-space
• Deep ReLU Networks Overcome the Curse of Dimensionality for Generalized Bandlimited Functions
• Deep Learning in High Dimension: Neural Network Expression Rates for Analytic Functions in \(\pmb{L^2(\mathbb{R}^d,\gamma_d)}\)
• Collocation approximation by deep neural ReLU networks for parametric and stochastic PDEs with lognormal inputs
• Approximation in shift-invariant spaces with deep ReLU neural networks

This page was built for publication: Deep ReLU neural network approximation in Bochner spaces and applications to parametric PDEs

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6062166)