Learning filter functions in regularisers by minimising quotients

DOI10.1007/978-3-319-58771-4_41zbMATH Open1489.68218arXiv1704.00989OpenAlexW2605133707MaRDI QIDQ5864036FDOQ5864036

Authors: Martin Benning, Guy Gilboa, Joana Sarah Grah, Carola-Bibiane Schönlieb 

Publication date: 3 June 2022

Published in: Lecture Notes in Computer Science (Search for Journal in Brave)

Abstract: Learning approaches have recently become very popular in the field of inverse problems. A large variety of methods has been established in recent years, ranging from bi-level learning to high-dimensional machine learning techniques. Most learning approaches, however, only aim at fitting parametrised models to favourable training data whilst ignoring misfit training data completely. In this paper, we follow up on the idea of learning parametrised regularisation functions by quotient minimisation as established in [3]. We extend the model therein to include higher-dimensional filter functions to be learned and allow for fit- and misfit-training data consisting of multiple functions. We first present results resembling behaviour of well-established derivative-based sparse regularisers like total variation or higher-order total variation in one-dimension. Our second and main contribution is the introduction of novel families of non-derivative-based regularisers. This is accomplished by learning favourable scales and geometric properties while at the same time avoiding unfavourable ones.

Full work available at URL: https://arxiv.org/abs/1704.00989

Recommendations

• Learning regularization parameters for general-form Tikhonov
• Regularization graphs -- a unified framework for variational regularization of inverse problems
• Learning nonlocal regularization operators
• A machine learning approach to optimal Tikhonov regularization. I: Affine manifolds
• A bilevel approach for parameter learning in inverse problems

zbMATH Keywords

sparse regularisationnonlinear eigenproblemgeneralised inverse power methodregularisation learning

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05) Image processing (compression, reconstruction, etc.) in information and communication theory (94A08) Regularity of solutions in optimal control (49N60) Inverse problems in optimal control (49N45)

Cited In (4)

• Nonlinear Power Method for Computing Eigenvectors of Proximal Operators and Neural Networks
• Modern regularization methods for inverse problems
• Rayleigh quotient minimization for absolutely one-homogeneous functionals
• Wasserstein-Based Projections with Applications to Inverse Problems

This page was built for publication: Learning filter functions in regularisers by minimising quotients

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5864036)