Uniform error estimates for artificial neural network approximations for heat equations

DOI10.1093/IMANUM/DRAB027MaRDI QIDQ5093099zbMATH OpenOpenAlexFDO

Authors Lukas Gonon, Arnulf Jentzen, David Kofler, David Šiška, Philipp Grohs

Publication date 26 July 2022

Published in IMA Journal of Numerical Analysis (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1911.09647

zbMATH Keywords

curse of dimensionalitydeep neural networkhigh-dimensional PDEsinformation based complexitytractability of multivariate problems

Mathematics Subject Classification ID

Heat equation (35K05) PDEs in connection with classical thermodynamics and heat transfer (35Q79) Computational methods for stochastic equations (aspects of stochastic analysis) (60H35) Probabilistic models, generic numerical methods in probability and statistics (65C20) Error bounds for initial value and initial-boundary value problems involving PDEs (65M15) Numerical methods for partial differential equations, initial value and time-dependent initial-boundary value problems (65M99) Complexity and performance of numerical algorithms (65Y20) Artificial neural networks and deep learning (68T07)

Abstract: Recently, artificial neural networks (ANNs) in conjunction with stochastic gradient descent optimization methods have been employed to approximately compute solutions of possibly rather high-dimensional partial differential equations (PDEs). Very recently, there have also been a number of rigorous mathematical results in the scientific literature which examine the approximation capabilities of such deep learning based approximation algorithms for PDEs. These mathematical results from the scientific literature prove in part that algorithms based on ANNs are capable of overcoming the curse of dimensionality in the numerical approximation of high-dimensional PDEs. In these mathematical results from the scientific literature usually the error between the solution of the PDE and the approximating ANN is measured in the $L^p$-sense with respect to some $pin[1,infty)$ and some probability measure. In many applications it is, however, also important to control the error in a uniform $L^{i}nfty$-sense. The key contribution of the main result of this article is to develop the techniques to obtain error estimates between solutions of PDEs and approximating ANNs in the uniform $L^{i}nfty$-sense. In particular, we prove that the number of parameters of an ANN to uniformly approximate the classical solution of the heat equation in a region $[a,b{]}^d$ for a fixed time point $Tin(0,infty)$ grows at most polynomially in the dimension $dinmathbbN$ and the reciprocal of the approximation precision $varepsilon>0$. This shows that ANNs can overcome the curse of dimensionality in the numerical approximation of the heat equation when the error is measured in the uniform $L^{i}nfty$-norm.

Recommendations

• A proof that rectified deep neural networks overcome the curse of dimensionality in the numerical approximation of semilinear heat equations
• Deep neural network approximations for solutions of PDEs based on Monte Carlo algorithms
• 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
• Error analysis for physics-informed neural networks (PINNs) approximating Kolmogorov PDEs
• Analysis of the generalization error: empirical risk minimization over deep artificial neural networks overcomes the curse of dimensionality in the numerical approximation of Black-Scholes partial differential equations

Cited in

(19)

• Deep neural network approximations for solutions of PDEs based on Monte Carlo algorithms
• A proof that rectified deep neural networks overcome the curse of dimensionality in the numerical approximation of semilinear heat equations
• Deep ReLU network expression rates for option prices in high-dimensional, exponential Lévy models
• Algorithms for solving high dimensional PDEs: from nonlinear Monte Carlo to machine learning
• Approximations with deep neural networks in Sobolev time-space
• Deep ReLU neural networks overcome the curse of dimensionality for partial integrodifferential equations
• Deep learning methods for partial differential equations and related parameter identification problems
• Overall error analysis for the training of deep neural networks via stochastic gradient descent with random initialisation
• Lower bounds for artificial neural network approximations: a proof that shallow neural networks fail to overcome the curse of dimensionality
• Numerical methods for backward stochastic differential equations: a survey
• Solving Kolmogorov PDEs without the curse of dimensionality via deep learning and asymptotic expansion with Malliavin calculus
• Approximation properties of residual neural networks for fractional differential equations
• Learning the random variables in Monte Carlo simulations with stochastic gradient descent: Machine learning for parametric PDEs and financial derivative pricing
• Deep neural network expressivity for optimal stopping problems
• The modified MSA, a gradient flow and convergence
• Overcoming the curse of dimensionality in the numerical approximation of high-dimensional semilinear elliptic partial differential equations
• Space-time deep neural network approximations for high-dimensional partial differential equations
• Universal approximation results for neural networks with non-polynomial activation function over non-compact domains
• Numerical analysis of residual neural networks for stochastic partial differential equations

This page was built for publication: Uniform error estimates for artificial neural network approximations for heat equations

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5093099)