Journal
SIAM JOURNAL ON IMAGING SCIENCES
Volume 14, Issue 1, Pages 92-127Publisher
SIAM PUBLICATIONS
DOI: 10.1137/20M1338460
Keywords
model correction; inverse problems; operator learning; deep learning; variational methods; photoacoustic tomography
Categories
Funding
- Academy of Finland [312123, 312342, 334817, 314411]
- British Heart Foundation [NH/18/1/33511]
- CMIC-EPSRC platform grant [EP/M020533/1]
- EPSRC-Wellcome grant [WT101957]
- Cambridge Centre for Analysis
- Cantab Capital Institute for the Mathematics of Information
- Leverhulme Trust
- EPSRC [EP/S026045/1, EP/T003553/1, EP/N022750/1, EP/T000864/1, EP/L016516/1]
- EPSRC Centre grant [EP/N014588/1]
- Wellcome Innovator Award [RG98755]
- RISE project CHiPS
- RISE project NoMADS
- Alan Turing Institute
- Jane and Aatos Erkko Foundation
- EPSRC [EP/T000864/1, EP/T003553/1, EP/M020533/1, EP/S026045/1, EP/N014588/1, EP/N022750/1] Funding Source: UKRI
- Academy of Finland (AKA) [334817, 334817] Funding Source: Academy of Finland (AKA)
Ask authors/readers for more resources
This paper discusses the potential of learning a data-driven explicit model correction for inverse problems within a variational framework. It presents a solution of forward-adjoint correction that explicitly corrects in both data and solution spaces, and derives conditions under which solutions to the variational problem with a learned correction converge to solutions obtained with the correct operator. The proposed approach is evaluated on limited view photoacoustic tomography and compared to the established Bayesian approximation error method, showing potential applications in this field.
We discuss the possibility of learning a data-driven explicit model correction for inverse problems and whether such a model correction can be used within a variational framework to obtain regularized reconstructions. This paper discusses the conceptual difficulty of learning such a forward model correction and proceeds to present a possible solution as a forward-adjoint correction that explicitly corrects in both data and solution spaces. We then derive conditions under which solutions to the variational problem with a learned correction converge to solutions obtained with the correct operator. The proposed approach is evaluated on an application to limited view photoacoustic tomography and compared to the established framework of the Bayesian approximation error method.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available