Error analysis for physics-informed neural networks (PINNs) approximating Kolmogorov PDEs
From MaRDI portal
Publication:2095545
DOI10.1007/S10444-022-09985-9zbMATH Open1502.65170arXiv2106.14473OpenAlexW3186608048MaRDI QIDQ2095545FDOQ2095545
Authors: Tim De Ryck, Siddhartha Mishra 
Publication date: 17 November 2022
Published in: Advances in Computational Mathematics (Search for Journal in Brave)
Abstract: Physics informed neural networks approximate solutions of PDEs by minimizing pointwise residuals. We derive rigorous bounds on the error, incurred by PINNs in approximating the solutions of a large class of linear parabolic PDEs, namely Kolmogorov equations that include the heat equation and Black-Scholes equation of option pricing, as examples. We construct neural networks, whose PINN residual (generalization error) can be made as small as desired. We also prove that the total -error can be bounded by the generalization error, which in turn is bounded in terms of the training error, provided that a sufficient number of randomly chosen training (collocation) points is used. Moreover, we prove that the size of the PINNs and the number of training samples only grow polynomially with the underlying dimension, enabling PINNs to overcome the curse of dimensionality in this context. These results enable us to provide a comprehensive error analysis for PINNs in approximating Kolmogorov PDEs.
Full work available at URL: https://arxiv.org/abs/2106.14473
Recommendations
- On the convergence of physics informed neural networks for linear second-order elliptic and parabolic type PDEs
- A Rate of Convergence of Physics Informed Neural Networks for the Linear Second Order Elliptic PDEs
- Error estimates and physics informed augmentation of neural networks for thermally coupled incompressible Navier Stokes equations
- Control of partial differential equations via physics-informed neural networks
- Physics informed neural networks (PINNs) for approximating nonlinear dispersive PDEs
Derivative securities (option pricing, hedging, etc.) (91G20) Artificial neural networks and deep learning (68T07) Heat equation (35K05) Nonlinear parabolic equations (35K55) KdV equations (Korteweg-de Vries equations) (35Q53) 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)
Cites Work
- fPINNs: Fractional Physics-Informed Neural Networks
- Stochastic differential equations. An introduction with applications.
- Title not available (Why is that?)
- Neural‐network‐based approximations for solving partial differential equations
- Title not available (Why is that?)
- On the approximation of functions by tanh neural networks
- Deep learning-based numerical methods for high-dimensional parabolic partial differential equations and backward stochastic differential equations
- Hidden physics models: machine learning of nonlinear partial differential equations
- Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
- Deep learning observables in computational fluid dynamics
- Deep learning in high dimension: neural network expression rates for generalized polynomial chaos expansions in UQ
- 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
- DeepXDE: a deep learning library for solving differential equations
- Physics-informed neural networks for high-speed flows
- Approximation rates for neural networks with encodable weights in smoothness spaces
- Iterative surrogate model optimization (ISMO): an active learning algorithm for PDE constrained optimization with deep neural networks
- Conservative physics-informed neural networks on discrete domains for conservation laws: applications to forward and inverse problems
- Error bounds for approximations with deep ReLU neural networks in \(W^{s , p}\) norms
- Extended Physics-Informed Neural Networks (XPINNs): A Generalized Space-Time Domain Decomposition Based Deep Learning Framework for Nonlinear Partial Differential Equations
- Error estimates for DeepONets: a deep learning framework in infinite dimensions
- B-PINNs: Bayesian physics-informed neural networks for forward and inverse PDE problems with noisy data
- Physics informed neural networks (PINNs) for approximating nonlinear dispersive PDEs
- Estimates on the generalization error of physics-informed neural networks for approximating a class of inverse problems for PDEs
Cited In (29)
- When do extended physics-informed neural networks (XPINNs) improve generalization?
- Uniform error estimates for artificial neural network approximations for heat equations
- Render unto numerics: orthogonal polynomial neural operator for PDEs with nonperiodic boundary conditions
- Error estimates for physics-informed neural networks approximating the Navier-Stokes equations
- Solving Poisson problems in polygonal domains with singularity enriched physics informed neural networks
- Boundary-safe PINNs extension: application to non-linear parabolic PDEs in counterparty credit risk
- Meshless methods for American option pricing through physics-informed neural networks
- Decoupling numerical method based on deep neural network for nonlinear degenerate interface problems
- Numerical analysis of physics-informed neural networks and related models in physics-informed machine learning
- Generalization of PINNs for elliptic interface problems
- Physics-informed neural networks for approximating dynamic (hyperbolic) PDEs of second order in time: error analysis and algorithms
- Mobility Estimation for Langevin Dynamics Using Control Variates
- Binary structured physics-informed neural networks for solving equations with rapidly changing solutions
- Recent developments in machine learning methods for stochastic control and games
- A short note on solving partial differential equations using convolutional neural networks
- A New Certified Hierarchical and Adaptive RB-ML-ROM Surrogate Model for Parametrized PDEs
- wPINNs: Weak Physics Informed Neural Networks for Approximating Entropy Solutions of Hyperbolic Conservation Laws
- Deep adaptive sampling for surrogate modeling without labeled data
- Error estimates and physics informed augmentation of neural networks for thermally coupled incompressible Navier Stokes equations
- Higher-order error estimates for physics-informed neural networks approximating the primitive equations
- Error estimates of physics-informed neural networks for initial value problems
- Error analysis of kernel/GP methods for nonlinear and parametric PDEs
- Reliable extrapolation of deep neural operators informed by physics or sparse observations
- On the convergence of physics informed neural networks for linear second-order elliptic and parabolic type PDEs
- Solving PDEs by variational physics-informed neural networks: an a posteriori error analysis
- Physics informed neural networks (PINNs) for approximating nonlinear dispersive PDEs
- Control of partial differential equations via physics-informed neural networks
- Greedy training algorithms for neural networks and applications to PDEs
- A theoretical case study of the generalization of machine-learned potentials
Uses Software
This page was built for publication: Error analysis for physics-informed neural networks (PINNs) approximating Kolmogorov PDEs
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2095545)
