Parametric encoding with attention and convolution mitigate spectral bias of neural partial differential equation solvers

Deep neural networks (DNNs) are increasingly used to solve partial differential equations (PDEs) that naturally arise while modeling a wide range of systems and physical phenomena. However, the accuracy of such DNNs decreases as the PDE complexity increases and they also suffer from spectral bias as...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	Structural and multidisciplinary optimization 2024-07, Vol.67 (7), p.128, Article 128
Hauptverfasser:	Shishehbor, Mehdi, Hosseinmardi, Shirin, Bostanabad, Ramin
Format:	Artikel
Sprache:	eng
Schlagworte:	Artificial neural networks Bias Coders Computational Mathematics and Numerical Analysis Convolution Domains Engineering Engineering Design Machine learning Partial differential equations Research Paper Theoretical and Applied Mechanics
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!