Abstract:In this work, we investigate the inverse problem of recovering a potential in an elliptic problem from random pointwise observations in the domain. We employ a regularized output-least squares formulation with an $H^1(\Omega)$ penalty for the numerical reconstruction, and the Galerkin finite element method for the spatial discretization. Under mild regularity assumptions on the problem data, we provide a stochastic $L^2(\Omega)$ convergence analysis on the regularized solution and the finite element approximation in a high probability sense. The obtained error bounds depend explicitly on the regularization parameter $\gamma$, the number $n$ of observation points and the mesh size $h$. These estimates provide a useful guideline for choosing relevant algorithmic parameters. Furthermore, we develop a monotonically convergent adaptive algorithm for determining a suitable regularization parameter in the absence of \textit{a priori} knowledge. Numerical experiments are provided to complement the theoretical results.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is the inverse problem of recovering the potential function $q(x)$ in the elliptic equation under random point - observed data. Specifically, the researchers focus on how to stably reconstruct a potential function $q(x)$ that appears as a coefficient in the elliptic equation from the random point - observed data within the domain $\Omega$. This problem has multiple practical applications, such as quantitative dynamic elastography (recovering density through one component of displacement), emissivity coefficient identification in heat conduction, and perfusion coefficient identification in the Pennes bio - heat conduction equation. ### Main contributions of the paper 1. **Almost optimal prior $\psi_2$-Orlicz bounds**: For the random regularized solution $q^*$ and the state approximation $u(q^\dagger)-u(q^*)$, when the regularization parameter $\gamma$ is appropriately selected (with a fixed noise level $\sigma$), almost optimal prior $\psi_2$-Orlicz bounds are provided. The specific bounds are as follows: \[ \left\| u(q^*) - u(q^\dagger) \right\|_{n,\psi_2}+\gamma^{1/2}\left\| q^* \right\|_{H^1(\Omega),\psi_2}\leq c\gamma^{1/2}\rho_0, \] where $\rho_0$ explicitly depends on $\left\| q^\dagger \right\|_{H^1(\Omega)}$, the noise level $\sigma$ and the number of observation points $n$. The key to the analysis lies in the peeling argument and the properties of the sub - Gaussian noise vector $\xi$. 2. **Stochastic $L^2(\Omega)$ error analysis of discrete solutions**: The Tikhonov functional and the control PDE are discretized using the continuous piecewise - linear finite - element method, and the theoretical results in the continuous setting are extended to the discrete setting, providing a stochastic $L^2(\Omega)$ error analysis of the discrete solution $q_h^*$ with high probability. 3. **Adaptive algorithm**: An adaptive algorithm is proposed to determine the appropriate regularization parameter $\gamma$. When no additional information is available, this algorithm only requires an initial value $\gamma_0$, and then iteratively updates $\gamma$ by solving the discrete regularization problem. The monotonic convergence of this adaptive algorithm is proved, and the convergence behaviors of the regularized solution and the adaptive algorithm are demonstrated through numerical experiments. ### Conclusion This paper makes important contributions to the stochastic convergence analysis of nonlinear PDE inverse problems, especially providing the stochastic convergence analysis of nonlinear PDE inverse problems for the first time. These results not only provide a theoretical basis for understanding and solving such inverse problems, but also provide guidance for parameter selection and algorithm design in practical applications.

Stochastic Convergence Analysis of Inverse Potential Problem

Convergence of Newton'S Method and Inverse Function Theorem in Banach Space

A Meshless Method for Some Inverse Problems Associated with the Helmholtz Equation

Stability for the Inverse Potential Problem by Finite Measurements on the Boundary

Stochastic Convergence of Regularized Solutions and Their Finite Element Approximations to Inverse Source Problems

A potential-free field inverse Schrödinger problem: optimal error bound analysis and regularization method

Stability of an inverse problem for the discrete wave equation and convergence results

Convergence of Numerov'S Method for Inverse Sturm-Liouville Problems

An inverse Cauchy problem of a stochastic hyperbolic equation

Stochastic Inverse Problem: stability, regularization and Wasserstein gradient flow

On a Modified Numerov's Method for Inverse Sturm-Liouville Problems

Convergence analysis of a stochastic heavy-ball method for linear ill-posed problems

Numerical Reconstruction of Diffusion and Potential Coefficients from Two Observations: Decoupled Recovery and Error Estimates

An inverse problem for a semi-linear wave equation: a numerical study

Inverse Sturm-Liouville problem with singular potential and spectral parameter in the boundary conditions

Recovering Simultaneously A Potential And A Point Source From Cauchy Data

Linearized Inverse Schrödinger Potential Problem with Partial Data and Its Deep Neural Network Inversion

Stochastic convergence of regularized solutions for backward heat conduction problems

Uniform stability of the inverse problem for the non-self-adjoint Sturm-Liouville operator

Numerical Recovery of a Time-Dependent Potential in Subdiffusion

Solving the inverse potential problem in the parabolic equation by the deep neural networks method