Abstract:We present a method for computing the inverse parameters and the solution field to inverse parametric PDEs based on randomized neural networks. This extends the local extreme learning machine technique originally developed for forward PDEs to inverse problems. We develop three algorithms for training the neural network to solve the inverse PDE problem. The first algorithm (NLLSQ) determines the inverse parameters and the trainable network parameters all together by the nonlinear least squares method with perturbations (NLLSQ-perturb). The second algorithm (VarPro-F1) eliminates the inverse parameters from the overall problem by variable projection to attain a reduced problem about the trainable network parameters only. It solves the reduced problem first by the NLLSQ-perturb algorithm for the trainable network parameters, and then computes the inverse parameters by the linear least squares method. The third algorithm (VarPro-F2) eliminates the trainable network parameters from the overall problem by variable projection to attain a reduced problem about the inverse parameters only. It solves the reduced problem for the inverse parameters first, and then computes the trainable network parameters afterwards. VarPro-F1 and VarPro-F2 are reciprocal to each other in a sense. The presented method produces accurate results for inverse PDE problems, as shown by the numerical examples herein. For noise-free data, the errors for the inverse parameters and the solution field decrease exponentially as the number of collocation points or the number of trainable network parameters increases, and can reach a level close to the machine accuracy. For noisy data, the accuracy degrades compared with the case of noise-free data, but the method remains quite accurate. The presented method has been compared with the physics-informed neural network method.

A randomized Newton's method for solving differential equations based on the neural network discretization

Gauss Newton method for solving variational problems of PDEs with neural network discretizaitons

Adaptive Growing Randomized Neural Networks for Solving Partial Differential Equations

Local Randomized Neural Networks with Discontinuous Galerkin Methods for Partial Differential Equations

A Method for Computing Inverse Parametric PDE Problems with Random-Weight Neural Networks

Neural Network As a Function Approximator and Its Application in Solving Differential Equations

Newton Informed Neural Operator for Computing Multiple Solutions of Nonlinear Partials Differential Equations

Fast Iterative Solver For Neural Network Method: II. 1D Diffusion-Reaction Problems And Data Fitting

Automatic Differentiation is Essential in Training Neural Networks for Solving Differential Equations

Neural-Network-Assisted Finite Difference Discretization for Numerical Solution of Partial Differential Equations

Solving partial differential equations with sampled neural networks

The Neural Network Approach to Inverse Problems in Differential Equations

A Priori Estimation of the Approximation, Optimization and Generalization Errors of Random Neural Networks for Solving Partial Differential Equations

Fast Iterative Solver For Neural Network Method: I. 1D Diffusion Problems

Subspace method based on neural networks for solving the partial differential equation

Accuracy and Architecture Studies of Residual Neural Network solving Ordinary Differential Equations

Solving Partial Differential Equations Using Artificial Neural Networks

Improved randomized neural network methods with boundary processing for solving elliptic equations

A Non-Gradient Method for Solving Elliptic Partial Differential Equations with Deep Neural Networks.

An Extreme Learning Machine-Based Method for Computational PDEs in Higher Dimensions