Learning nonlocal regularization operators
Mathematical Control & Related Fields · 2022 · 12 (1) · 81
DOI: 10.3934/mcrf.2021003 ↗Kurz gefasst
Fachartikel in der Zeitschrift Mathematical Control & Related Fields, 2022. Verfasst von Gernot Holler und 1 weiteren Autor:innen. Die Arbeit wurde bisher 11-mal zitiert (Crossref) und nennt 31 Literaturangaben.
Abstract (Originalsprache)
A learning approach for determining which operator from a class of nonlocal operators is optimal for the regularization of an inverse problem is investigated. The considered class of nonlocal operators is motivated by the use of squared fractional order Sobolev seminorms as regularization operators. First fundamental results from the theory of regularization with local operators are extended to the nonlocal case. Then a framework based on a bilevel optimization strategy is developed which allows to choose nonlocal regularization operators from a given class which i) are optimal with respect to a suitable performance measure on a training set, and ii) enjoy particularly favorable properties. Results from numerical experiments are also provided.
Zitieren
APA
Holler, G., & Kunisch, K. (2022). Learning nonlocal regularization operators. Mathematical Control & Related Fields, 12(1), 81. https://doi.org/10.3934/mcrf.2021003
Vancouver
Holler G, Kunisch K. Learning nonlocal regularization operators. Mathematical Control & Related Fields. 2022;12(1):81. doi:10.3934/mcrf.2021003
BibTeX
@article{Holler2022,
author = {Holler, Gernot and Kunisch, Karl},
title = {Learning nonlocal regularization operators},
journal = {Mathematical Control & Related Fields},
year = {2022},
volume = {12},
number = {1},
pages = {81},
doi = {10.3934/mcrf.2021003}
}
EUVIMED