Continuous limits of residual neural networks in case of large input data Cover

.blurhash-client-img { display: none !important; }

Continuous limits of residual neural networks in case of large input data

Communications in Applied and Industrial Mathematics

Volume 13 (2022): Issue 1 (January 2022)

By: Michael Herty, Anna Thünen, Torsten Trimborn and Giuseppe Visconti

Open Access

|Dec 2022

Authors

Michael Herty

herty@igpm.rwth-aachen.de

Institut für Geometrie und Praktische Mathematik (IGPM), RWTH Aachen University, Aachen, Germany

Anna Thünen

Institut für Mathematik – TU Clausthal – Erzstraße 1, 38678 Clausthal-Zellerfeld, Germany

Torsten Trimborn

NRW.BANK – Kavalleriestraße 22, 40213 Düsseldorf, Germany

Giuseppe Visconti

Department of Mathematics, “G. Castelnuovo” – Sapienza University of Rome, Roma, Italy

DOI: https://doi.org/10.2478/caim-2022-0008 | Journal eISSN: 2038-0909

Journal RSS Feed

Language: English

Page range: 96 - 120

Submitted on: Jul 11, 2022

|

Accepted on: Nov 12, 2022

|

Published on: Dec 24, 2022

Published by: Italian Society for Applied and Industrial Mathemathics

In partnership with: Paradigm Publishing Services

Publication frequency: 1 issue per year

Keywords:

Neural networks,

mean-field limit,

well-posedness,

optimal control,

controllability

Related subjects:

Numerical and computational mathematics,

Applied mathematics

© 2022 Michael Herty, Anna Thünen, Torsten Trimborn, Giuseppe Visconti, published by Italian Society for Applied and Industrial Mathemathics
This work is licensed under the Creative Commons Attribution 4.0 License.

Previous article Volume 13 (2022): Issue 1 (January 2022)