HomPINNs: homotopy physics-informed neural networks for solving the inverse problems of nonlinear differential equations with multiple solutions

DOI10.1016/j.jcp.2023.112751arXiv2304.02811MaRDI QIDQ6119293

Ziyang Huang, Haoyang Zheng, Wenrui Hao, Yao Huang, Guang Lin

Publication date: 29 February 2024

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

Full work available at URL: https://arxiv.org/abs/2304.02811

zbMATH Keywords

nonlinear differential equationsinverse problemsmultiple solutionsmachine learninghomotopy continuation methodphysics-informed neural networks

Mathematics Subject Classification ID

Artificial intelligence (68Txx) Numerical methods for partial differential equations, boundary value problems (65Nxx) Partial differential equations of mathematical physics and other areas of application (35Qxx)

Cites Work

• A homotopy method based on WENO schemes for solving steady state problems of hyperbolic conservation laws
• A physics-informed deep learning framework for inversion and surrogate modeling in solid mechanics
• B-PINNs: Bayesian physics-informed neural networks for forward and inverse PDE problems with noisy data
• NSFnets (Navier-Stokes flow nets): physics-informed neural networks for the incompressible Navier-Stokes equations
• Thermodynamically consistent physics-informed neural networks for hyperbolic systems
• When and why PINNs fail to train: a neural tangent kernel perspective
• HomPINNs: Homotopy physics-informed neural networks for learning multiple solutions of nonlinear elliptic differential equations
• Conservative physics-informed neural networks on discrete domains for conservation laws: applications to forward and inverse problems
• A tutorial on the adjoint method for inverse problems
• On the eigenvector bias of Fourier feature networks: from regression to solving multi-scale PDEs with physics-informed neural networks
• A bootstrapping approach for computing multiple solutions of differential equations
• A homotopy method with adaptive basis selection for computing multiple solutions of differential equations
• Spatial pattern formation in reaction-diffusion models: a computational approach
• An adaptive homotopy method for computing bifurcations of nonlinear parametric systems
• Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
• Ensemble Kalman methods for inverse problems
• The Cheater’s Homotopy: An Efficient Procedure for Solving Systems of Polynomial Equations
• Rotating Chemical Waves in the Gray–Scott Model
• Neural‐network‐based approximations for solving partial differential equations
• Hidden fluid mechanics: Learning velocity and pressure fields from flow visualizations
• DeepXDE: A Deep Learning Library for Solving Differential Equations
• Extended Physics-Informed Neural Networks (XPINNs): A Generalized Space-Time Domain Decomposition Based Deep Learning Framework for Nonlinear Partial Differential Equations
• fPINNs: Fractional Physics-Informed Neural Networks
• A Computational Framework for Infinite-Dimensional Bayesian Inverse Problems Part I: The Linearized Case, with Application to Global Seismic Inversion

This page was built for publication: HomPINNs: homotopy physics-informed neural networks for solving the inverse problems of nonlinear differential equations with multiple solutions