Abstract:We study a family of $H^m$-conforming piecewise polynomials based on artificial neural network, named as the finite neuron method (FNM), for numerical solution of $2m$-th order partial differential equations in $\mathbb{R}^d$ for any $m,d \geq 1$ and then provide convergence analysis for this method. Given a general domain $\Omega\subset\mathbb R^d$ and a partition $\mathcal T_h$ of $\Omega$, it is still an open problem in general how to construct conforming finite element subspace of $H^m(\Omega)$ that have adequate approximation properties. By using techniques from artificial neural networks, we construct a family of $H^m$-conforming set of functions consisting of piecewise polynomials of degree $k$ for any $k\ge m$ and we further obtain the error estimate when they are applied to solve elliptic boundary value problem of any order in any dimension. For example, the following error estimates between the exact solution $u$ and finite neuron approximation $u_N$ are obtained. $$ \|u-u_N\|_{H^m(\Omega)}=\mathcal O(N^{-{1\over 2}-{1\over d}}). $$ Discussions will also be given on the difference and relationship between the finite neuron method and finite element methods (FEM). For example, for finite neuron method, the underlying finite element grids are not given a priori and the discrete solution can only be obtained by solving a non-linear and non-convex optimization problem. Despite of many desirable theoretical properties of the finite neuron method analyzed in the paper, its practical value is a subject of further investigation since the aforementioned underlying non-linear and non-convex optimization problem can be expensive and challenging to solve. For completeness and also convenience to readers, some basic known results and their proofs are also included in this manuscript.

Finite element analysis of extended Fisher-Kolmogorov equation with Neumann boundary conditions

Finite Element Solution in Fluid Mechanics Using Exponential Function Based Interpolation

Superconvergence analysis of a low order expanded conforming mixed finite element method for the extended Fisher–Kolmogorov equation

A General Finite Element Method: Extension of Variational Analysis for Nonlinear Heat Conduction with Temperature-Dependent Properties and Boundary Conditions, and Its Implementation As Local Refinement.

Numerical solution and analysis of extended Fisher–Kolmogorov equation using an improved collocation algorithm

Analysis and simulation of the (2 + 1)-dimensional Fisher's reaction-diffusion equation with higher order finite element method

An adaptive low-rank splitting approach for the extended Fisher–Kolmogorov equation

A hybrid kernel-based meshless method for numerical approximation of multidimensional Fisher's equation

FDM/FEM for nonlinear convection–diffusion–reaction equations with Neumann boundary conditions—Convergence analysis for smooth and nonsmooth solutions

Unconditional superconvergence analysis of two-grid nonconforming FEMs for the fourth order nonlinear extend Fisher-Kolmogorov equation

Finite element approximation of stationary Fokker--Planck--Kolmogorov equations with application to periodic numerical homogenization

A new 2-level implicit high accuracy compact exponential approximation for the numerical solution of nonlinear fourth order Kuramoto–Sivashinsky and Fisher–Kolmogorov equations

The dual reciprocity boundary elements method for one-dimensional nonlinear parabolic partial differential equations

Finite Element Approximation of Hamilton-Jacobi-Bellman equations with nonlinear mixed boundary conditions

The Finite Neuron Method and Convergence Analysis

ON SPECIAL FINITE DIFFERENCE APPROXIMATIONS FOR SOLVING SECOND ORDER DIFFERENTIAL EQUATIONS

On full linear convergence and optimal complexity of adaptive FEM with inexact solver

Finite element approximation for uniformly elliptic linear PDE of second order in nondivergence form

Stability and Convergence Analysis of an Exact Finite Difference Scheme for Fredholm Integro-Differential Equations

Solving nonlinear Fisher–Kolmogorov–Petrovsky–Piskunov equation using two meshless methods

A spectral-based numerical method for Kolmogorov equations in Hilbert spaces