Isogeometric neural networks: a new deep learning approach for solving parameterized partial differential equations

DOI10.1016/j.cma.2022.115839OpenAlexW4313334586MaRDI QIDQ2683423

Joshua Gasick, Xiao-ping Qian

Publication date: 10 February 2023

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

Full work available at URL: https://doi.org/10.1016/j.cma.2022.115839

zbMATH Keywords

partial differential equationsunsupervised learningisogeometric analysisNURBSphysics-informed learning

Mathematics Subject Classification ID

Numerical analysis (65-XX) Computer science (68-XX)

Related Items (1)

Physically informed deep homogenization neural network for unidirectional multiphase/multi-inclusion thermoconductive composites

Uses Software

• L-BFGS
• TensorFlow
• DiffSharp
• DeepXDE
• SciANN
• DGM

Cites Work

• Unnamed Item
• Unnamed Item
• Isogeometric analysis: CAD, finite elements, NURBS, exact geometry and mesh refinement
• On the limited memory BFGS method for large scale optimization
• Solution of nonlinear ordinary differential equations by feedforward neural networks
• Inferring solutions of differential equations using noisy multi-fidelity data
• Multilayer feedforward networks are universal approximators
• DGM: a deep learning algorithm for solving partial differential equations
• SciANN: a Keras/Tensorflow wrapper for scientific computations and physics-informed deep learning using artificial neural networks
• \textit{hp}-VPINNs: variational physics-informed neural networks with domain decomposition
• A physics-informed deep learning framework for inversion and surrogate modeling in solid mechanics
• A method for representing periodic functions and enforcing exactly periodic boundary conditions with deep neural networks
• B-PINNs: Bayesian physics-informed neural networks for forward and inverse PDE problems with noisy data
• PhyGeoNet: physics-informed geometry-adaptive convolutional neural networks for solving parameterized steady-state PDEs on irregular domain
• When and why PINNs fail to train: a neural tangent kernel perspective
• Surrogate modeling for fluid flows based on physics-constrained deep learning without simulation data
• Physics-constrained deep learning for high-dimensional surrogate modeling and uncertainty quantification without labeled data
• Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
• Conditional deep surrogate models for stochastic, high-dimensional, and multi-fidelity systems
• Understanding and Mitigating Gradient Flow Pathologies in Physics-Informed Neural Networks
• Numerical solution of inverse problems by weak adversarial networks
• DeepXDE: A Deep Learning Library for Solving Differential Equations
• On the Convergence of Physics Informed Neural Networks for Linear Second-Order Elliptic and Parabolic Type PDEs
• Multi-Scale Deep Neural Network (MscaleDNN) Methods for Oscillatory Stokes Flows in Complex Domains
• ISOGEOMETRIC ANALYSIS: APPROXIMATION, STABILITY AND ERROR ESTIMATES FOR h-REFINED MESHES

This page was built for publication: Isogeometric neural networks: a new deep learning approach for solving parameterized partial differential equations