A hybrid deep neural operator/finite element method for ice-sheet modeling

DOI10.1016/J.JCP.2023.112428arXiv2301.11402MaRDI QIDQ6054205

Amanda A. Howard, George Em. Karniadakis, Panos Stinis, Mauro Perego, Qizhi He

Publication date: 27 September 2023

Published in: Journal of Computational Physics (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/2301.11402

zbMATH Keywords

finite elementhybrid modelneural operatorice-sheet dynamicsdeep learning surrogate

Mathematics Subject Classification ID

Basic methods in fluid mechanics (76Mxx) Probabilistic methods, stochastic differential equations (65Cxx) Geophysics (86Axx)

Related Items (1)

Multifidelity deep operator networks for data-driven and physics-informed problems

Cites Work

• Numerical methods for the discretization of random fields by means of the Karhunen-Loève expansion
• Scalable and efficient algorithms for the propagation of uncertainty from data through inference to prediction for large-scale problems, with application to flow of the antarctic ice sheet
• Theory of shallow ice shelves
• Self-adaptive physics-informed neural networks
• Variational inference at glacier scale
• Optimal Model Management for Multifidelity Monte Carlo Estimation
• Universal Approximation of Multiple Nonlinear Operators by Neural Networks
• A seamless multiscale operator neural network for inferring bubble dynamics
• Mechanical error estimators for shallow ice flow models

This page was built for publication: A hybrid deep neural operator/finite element method for ice-sheet modeling