Sampled Limited Memory Methods for Massive Linear Inverse Problems

TL;DR

This paper introduces a sampled limited memory row-action method for large linear inverse problems, improving convergence speed and accuracy in high-dimensional imaging applications like tomography.

Contribution

It proposes a novel method that generalizes the damped block Kaczmarz approach, incorporating curvature approximation to enhance convergence in massive inverse problems.

Findings

Method accelerates convergence in large-scale tomography problems.

Linear convergence proven for the expectation of iterates.

Numerical results show improved accuracy and efficiency.

Abstract

In many modern imaging applications the desire to reconstruct high resolution images, coupled with the abundance of data from acquisition using ultra-fast detectors, have led to new challenges in image reconstruction. A main challenge is that the resulting linear inverse problems are massive. The size of the forward model matrix exceeds the storage capabilities of computer memory, or the observational dataset is enormous and not available all at once. Row-action methods that iterate over samples of rows can be used to approximate the solution while avoiding memory and data availability constraints. However, their overall convergence can be slow. In this paper, we introduce a sampled limited memory row-action method for linear least squares problems, where an approximation of the global curvature of the underlying least squares problem is used to speed up the initial convergence and to improve the accuracy of iterates. We show that this limited memory method is a generalization of the damped block Kaczmarz method, and we prove linear convergence of the expectation of the iterates and of the error norm up to a convergence horizon. Numerical experiments demonstrate the benefits of these sampled limited memory row-action methods for massive 2D and 3D inverse problems in tomography applications.