11
11

Consistency of Bayesian inference with Gaussian process priors for a parabolic inverse problem

Abstract

We consider the statistical nonlinear inverse problem of recovering the absorption term f>0f>0 in the heat equation \partial_tu-\frac{1}{2}\Delta u+fu=0 \quad \text{on O×(0,T)\mathcal{O}\times(0,\textbf{T})}\quad u = g \quad \text{on O×(0,T)\partial\mathcal{O}\times(0,\textbf{T})}\quad u(\cdot,0)=u_0 \quad \text{on O\mathcal{O}}, where ORd\mathcal{O}\in\mathbb{R}^d is a bounded domain, T<\textbf{T}<\infty is a fixed time, and g,u0g,u_0 are given sufficiently smooth functions describing boundary and initial values respectively. The data consists of NN discrete noisy point evaluations of the solution ufu_f on O×(0,T)\mathcal{O}\times(0,\textbf{T}). We study the statistical performance of Bayesian nonparametric procedures based on a large class of Gaussian process priors. We show that, as the number of measurements increases, the resulting posterior distributions concentrate around the true parameter generating the data, and derive a convergence rate for the reconstruction error of the associated posterior means. We also consider the optimality of the contraction rates and prove a lower bound for the minimax convergence rate for inferring ff from the data, and show that optimal rates can be achieved with truncated Gaussian priors.

View on arXiv
Comments on this paper