Doorgaan naar hoofdnavigatie Doorgaan naar zoeken Ga verder naar hoofdinhoud

Improved Generalization of Symplectic Neural Networks for Hamiltonian Systems

Onderzoeksoutput: Hoofdstuk in Boek/Rapport/CongresprocedureConferentiebijdrageAcademicpeer review

Samenvatting

We analyze the impact of structure preservation inside the topology of a neural network on the error of the neural network. To do this we split the error into an approximation, optimization and generalization term. The analysis is of empirical nature and is performed on the example of preserving the symplectic structure of Hamiltonian systems. For this, three different types of neural networks are trained on data from two different Hamiltonian systems. We find that the main improvement of structure-preserving neural networks comes in the form of a greatly reduced generalization error.

Originele taal-2Engels
TitelNumerical Mathematics and Advanced Applications ENUMATH 2023, Volume 1
SubtitelEuropean Conference, September 4-8, Lisbon, Portugal
RedacteurenAdélia Sequeira, Ana Silvestre, Svilen S. Valtchev, João Janela
Plaats van productieCham
UitgeverijSpringer
Pagina's465-472
Aantal pagina's8
ISBN van elektronische versie978-3-031-86173-4
ISBN van geprinte versie978-3-031-86172-7, 978-3-031-86175-8
DOI's
StatusGepubliceerd - 25 apr. 2025
EvenementEuropean Conference on Numerical Mathematics and Advanced Applications, ENUMATH 2023 - Lisbon, Portugal
Duur: 4 sep. 20238 sep. 2023

Publicatie series

NaamLecture Notes in Computational Science and Engineering (LNCSE)
Volume153
ISSN van geprinte versie1439-7358
ISSN van elektronische versie2197-7100

Congres

CongresEuropean Conference on Numerical Mathematics and Advanced Applications, ENUMATH 2023
Land/RegioPortugal
StadLisbon
Periode4/09/238/09/23

Vingerafdruk

Duik in de onderzoeksthema's van 'Improved Generalization of Symplectic Neural Networks for Hamiltonian Systems'. Samen vormen ze een unieke vingerafdruk.

Citeer dit