English

Efficient learning methods for large-scale optimal inversion design

Numerical Analysis 2021-10-07 v1 Numerical Analysis

Abstract

In this work, we investigate various approaches that use learning from training data to solve inverse problems, following a bi-level learning approach. We consider a general framework for optimal inversion design, where training data can be used to learn optimal regularization parameters, data fidelity terms, and regularizers, thereby resulting in superior variational regularization methods. In particular, we describe methods to learn optimal pp and qq norms for LpLq{\rm L}^p-{\rm L}^q regularization and methods to learn optimal parameters for regularization matrices defined by covariance kernels. We exploit efficient algorithms based on Krylov projection methods for solving the regularized problems, both at training and validation stages, making these methods well-suited for large-scale problems. Our experiments show that the learned regularization methods perform well even when there is some inexactness in the forward operator, resulting in a mixture of model and measurement error.

Keywords

Cite

@article{arxiv.2110.02720,
  title  = {Efficient learning methods for large-scale optimal inversion design},
  author = {Julianne Chung and Matthias Chung and Silvia Gazzola and Mirjeta Pasha},
  journal= {arXiv preprint arXiv:2110.02720},
  year   = {2021}
}