Relaxed Gauss-Newton methods with applications to electrical impedance tomography

Abstract

As second-order methods, Gauss--Newton-type methods can be more effective than first-order methods for the solution of nonsmooth optimization problems with expensive-to-evaluate smooth components. Such methods, however, often do not converge. Motivated by nonlinear inverse problems with nonsmooth regularization, we propose a new Gauss--Newton-type method with inexact relaxed steps. We prove that the method converges to a set of disjoint critical points given that the linearisation of the forward operator for the inverse problem is sufficiently precise. We extensively evaluate the performance of the method on electrical impedance tomography (EIT).

arXiv comment43 pages, 29 figures

References in corpus (3)

• A primal-dual hybrid gradient method for non-linear operators with applications to MRI
• Acceleration and global convergence of a first-order primal--dual method for nonconvex problems
• Block-proximal methods with spatially adapted acceleration

Cited by in corpus (9)

• Non-planar sensing skins for structural health monitoring based on electrical resistance tomography
• First-order primal-dual methods for nonsmooth nonconvex optimisation
• Mumford-Shah regularization in electrical impedance tomography with complete electrode model
• Electrodeless electrode model for electrical impedance tomography
• Regularisation, optimisation, subregularity
• A nonsmooth primal-dual method with interwoven PDE constraint solver
• Predictive online optimisation with applications to optical flow
• Solver-in-the-Loop joint operator learning: fractional Laplace-Beltrami features for interface reconstruction
• Online optimisation for dynamic electrical impedance tomography