A parallel sampling algorithm for inverse problems with linear and nonlinear unknowns

TL;DR

This paper introduces a parallel sampling algorithm for inverse problems involving both linear and nonlinear unknowns, effectively handling noisy and ill-posed data, with applications demonstrated in seismology.

Contribution

The paper presents a novel parallel sampling method that jointly estimates linear and nonlinear unknowns without limiting to maximum likelihood values, improving computational efficiency.

Findings

Algorithm performs well on seismic inverse problems

Results compare favorably to ML, GCV, and CLS methods

Parallel approach reduces computational cost

Abstract

We derive a parallel sampling algorithm for computational inverse problems that present an unknown linear forcing term and a vector of nonlinear parameters to be recovered. It is assumed that the data is noisy and that the linear part of the problem is ill-posed. The vector of nonlinear parameters m is modeled as a random variable. A dilation parameter alpha is used to scale the regularity of the linear unknown and is also modeled as a random variable. A posterior probability distribution for (m; alpha) is derived following an approach related to the maximum likelihood regularization parameter selection [5]. A major difference in our approach is that, unlike in [5], we do not limit ourselves to the maximum likelihood value of alpha. We then derive a parallel sampling algorithm where we alternate computing proposals in parallel and combining proposals to accept or reject them as in [4]. This algorithm is well-suited to problems where proposals are expensive to compute. We then apply it to an inverse problem in seismology. We show how our results compare favorably to those obtained from the Maximum Likelihood (ML), the Generalized Cross Validation (GCV), and the Constrained Least Squares (CLS) algorithms.