Abstract:Throughout many fields, practitioners often rely on differential equations to model systems. Yet, for many applications, the theoretical derivation of such equations and/or accurate resolution of their solutions may be intractable. Instead, recently developed methods, including those based on parameter estimation, operator subset selection, and neural networks, allow for the data-driven discovery of both ordinary and partial differential equations (PDEs), on a spectrum of interpretability. The success of these strategies is often contingent upon the correct identification of representative equations from noisy observations of state variables and, as importantly and intertwined with that, the mathematical strategies utilized to enforce those equations. Specifically, the latter has been commonly addressed via unconstrained optimization strategies. Representing the PDE as a neural network, we propose to discover the PDE by solving a constrained optimization problem and using an intermediate state representation similar to a Physics-Informed Neural Network (PINN). The objective function of this constrained optimization problem promotes matching the data, while the constraints require that the PDE is satisfied at several spatial collocation points. We present a penalty method and a widely used trust-region barrier method to solve this constrained optimization problem, and we compare these methods on numerical examples. Our results on the Burgers' and the Korteweg-De Vreis equations demonstrate that the latter constrained method outperforms the penalty method, particularly for higher noise levels or fewer collocation points. For both methods, we solve these discovered neural network PDEs with classical methods, such as finite difference methods, as opposed to PINNs-type methods relying on automatic differentiation. We briefly highlight other small, yet crucial, implementation details.

Discovery of PDEs Driven by Data with Sharp Gradient or Discontinuity

DL-PDE: Deep-learning based data-driven discovery of partial differential equations from discrete and noisy data

Physics-Guided Discovery of Highly Nonlinear Parametric Partial Differential Equations

Deep-learning of parametric partial differential equations from sparse and noisy data

Robust data-driven discovery of governing physical laws with error bars

Deep-learning based discovery of partial differential equations in integral form from sparse and noisy data

Data-Driven Discovery of PDEs via the Adjoint Method

DLGA-PDE: Discovery of PDEs with incomplete candidate library via combination of deep learning and genetic algorithm

Bayesian data-driven discovery of partial differential equations with variable coefficients

Data-driven discovery of stochastic differential equations

Discovery of partial differential equations from highly noisy and sparse data with physics-informed information criterion

Learning PDEs from data on closed surfaces with sparse optimization

A Geometric Perspective on Diffusion Models

DiffusionPDE: Generative PDE-Solving Under Partial Observation

Discontinuity Computing using Physics-Informed Neural Network

Gradient estimates for parabolic nonlinear nonlocal equations

Constrained or Unconstrained? Neural-Network-Based Equation Discovery from Data

Enhancement of shock-capturing methods via machine learning

Data-driven discovery of Green's functions

Convergence to Sharp Traveling Waves of Solutions for Burgers-Fisher-KPP Equations with Degenerate Diffusion

A Sharp Convergence Theory for The Probability Flow ODEs of Diffusion Models