Separable physics-informed DeepONet: breaking the curse of dimensionality in physics-informed machine learning

DOI10.1016/J.CMA.2024.117586MaRDI QIDQ6669073FDOQ6669073

Authors: Luis Mandl, Somdatta Goswami, Tim Ricken 

Publication date: 22 January 2025

Published in: Computer Methods in Applied Mechanics and Engineering (Search for Journal in Brave)

Recommendations

• A dimension-augmented physics-informed neural network (DaPINN) with high level accuracy and efficiency
• Navigating PINNs via maximum residual-based continuous distribution
• Hutchinson trace estimation for high-dimensional and high-order physics-informed neural networks
• Diminishing spectral bias in physics-informed neural networks using spatially-adaptive Fourier feature encoding

zbMATH Keywords

separation of variablesscientific machine learningphysics-informedneural operator

Mathematics Subject Classification ID

Statistical mechanics, structure of matter (82-XX) Quantum theory (81-XX)

Cites Work

• Tensor Rank and the Ill-Posedness of the Best Low-Rank Approximation Problem
• A tensor approximation method based on ideal minimal residual formulations for the solution of high-dimensional problems
• Title not available (Why is that?)
• Understanding and Mitigating Gradient Flow Pathologies in Physics-Informed Neural Networks
• Phase-field DeepONet: physics-informed deep operator neural network for fast simulations of pattern formation governed by gradient flows of free-energy functionals
• A physics-informed variational DeepONet for predicting crack path in quasi-brittle materials
• A comprehensive and fair comparison of two neural operators (with practical extensions) based on FAIR data
• Multifidelity deep operator networks for data-driven and physics-informed problems
• General theory of three-dimensional consolidation.
• Wavelet neural operator for solving parametric partial differential equations in computational mechanics problems
• PGD reduced-order modeling for structural dynamics applications
• On the influence of over-parameterization in manifold based surrogates and deep neural operators
• MIONet: Learning Multiple-Input Operators via Tensor Product
• Respecting causality for training physics-informed neural networks
• Energy-dissipative evolutionary deep operator neural networks
• Zero coordinate shift: whetted automatic differentiation for physics-informed operator learning
• SimLivA-modeling ischemia-reperfusion injury in the liver: a first step towards a clinical decision support tool

Cited In (1)

• Hutchinson trace estimation for high-dimensional and high-order physics-informed neural networks

This page was built for publication: Separable physics-informed DeepONet: breaking the curse of dimensionality in physics-informed machine learning

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6669073)