Geodesic
Deep learning for gradient flows using the Brezis–Ekeland principle
Archivum mathematicum, Tome 59 (2023) no. 3, pp. 249-261.

Voir la notice de l'article provenant de la source Czech Digital Mathematics Library

We propose a deep learning method for the numerical solution of partial differential equations that arise as gradient flows. The method relies on the Brezis–Ekeland principle, which naturally defines an objective function to be minimized, and so is ideally suited for a machine learning approach using deep neural networks. We describe our approach in a general framework and illustrate the method with the help of an example implementation for the heat equation in space dimensions two to seven.
DOI : 10.5817/AM2023-3-249
Classification : 35A15, 35K15, 68t07
Keywords: machine learning; deep neural networks; gradient flows; Brezis–Ekeland principle; adversarial networks; differential equations 
@article{10_5817_AM2023_3_249,
     author = {Carini, Laura and Jensen, Max and N\"urnberg, Robert},
     title = {Deep learning for gradient flows using the {Brezis{\textendash}Ekeland} principle},
     journal = {Archivum mathematicum},
     pages = {249--261},
     publisher = {mathdoc},
     volume = {59},
     number = {3},
     year = {2023},
     doi = {10.5817/AM2023-3-249},
     mrnumber = {4563037},
     zbl = {07675595},
     language = {en},
     url = {http://geodesic.mathdoc.fr/articles/10.5817/AM2023-3-249/}
}
TY  - JOUR
AU  - Carini, Laura
AU  - Jensen, Max
AU  - Nürnberg, Robert
TI  - Deep learning for gradient flows using the Brezis–Ekeland principle
JO  - Archivum mathematicum
PY  - 2023
SP  - 249
EP  - 261
VL  - 59
IS  - 3
PB  - mathdoc
UR  - http://geodesic.mathdoc.fr/articles/10.5817/AM2023-3-249/
DO  - 10.5817/AM2023-3-249
LA  - en
ID  - 10_5817_AM2023_3_249
ER  - 
%0 Journal Article
%A Carini, Laura
%A Jensen, Max
%A Nürnberg, Robert
%T Deep learning for gradient flows using the Brezis–Ekeland principle
%J Archivum mathematicum
%D 2023
%P 249-261
%V 59
%N 3
%I mathdoc
%U http://geodesic.mathdoc.fr/articles/10.5817/AM2023-3-249/
%R 10.5817/AM2023-3-249
%G en
%F 10_5817_AM2023_3_249
Carini, Laura; Jensen, Max; Nürnberg, Robert. Deep learning for gradient flows using the Brezis–Ekeland principle. Archivum mathematicum, Tome 59 (2023) no. 3, pp. 249-261. doi : 10.5817/AM2023-3-249. http://geodesic.mathdoc.fr/articles/10.5817/AM2023-3-249/

Cité par Sources :