CEENs: causality-enforced evolutional networks for solving time-dependent partial differential equations

DOI10.1016/J.CMA.2024.117036MaRDI QIDQ6557798zbMATH OpenarXivFDO

Authors Heechang Kim, Minseok Choi, Jeahan Jung, Hyomin Shin

Publication date 18 June 2024

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

Tldr We propose CEENs: Causality-enforced evolutional networks, which integrate temporal causality into physics-informed neural networks (PINNs) by dividing the time domain into nonoverlapping subintervals and assigning unique neural networks to each, sequentially training them while enforcing causality through a novel loss function founded on the integral form of PDEs within these intervals.

Full work available at URL https://doi.org/10.1016/J.CMA.2024.117036

zbMATH Keywords

long-time integrationnonlinear dynamicspartial differential equationphysics-informed neural networkpredictive modeling

Mathematics Subject Classification ID

Finite difference methods for initial value and initial-boundary value problems involving PDEs (65M06) Spectral, collocation and related methods for initial value and initial-boundary value problems involving PDEs (65M70)

arXiv classification

cs.LGmath.NA

Recommendations

• Exact enforcement of temporal continuity in sequential physics-informed neural networks
• Pre-training strategy for solving evolution equations based on physics-informed neural networks
• PhyCRNet: physics-informed convolutional-recurrent network for solving spatiotemporal PDEs
• A novel sequential method to train physics informed neural networks for Allen Cahn and Cahn Hilliard equations
• Gradient-enhanced physics-informed neural networks for forward and inverse PDE problems

Cites work

• A composite neural network that learns from multi-fidelity data: application to function approximation and inverse PDE problems
• A cusp-capturing PINN for elliptic interface problems
• A method for representing periodic functions and enforcing exactly periodic boundary conditions with deep neural networks
• A novel sequential method to train physics informed neural networks for Allen Cahn and Cahn Hilliard equations
• Adaptive activation functions accelerate convergence in deep and physics-informed neural networks
• B-PINNs: Bayesian physics-informed neural networks for forward and inverse PDE problems with noisy data
• Bayesian Deep Learning Framework for Uncertainty Quantification in Stochastic Partial Differential Equations
• Conservative physics-informed neural networks on discrete domains for conservation laws: applications to forward and inverse problems
• Learning in Modal Space: Solving Time-Dependent Stochastic PDEs Using Physics-Informed Neural Networks
• Learning time-dependent PDEs with a linear and nonlinear separate convolutional neural network
• Local extreme learning machines and domain decomposition for solving linear and nonlinear partial differential equations
• Modeling the dynamics of PDE systems with physics-constrained deep auto-regressive networks
• Neural‐network‐based approximations for solving partial differential equations
• Normalizing field flows: solving forward and inverse stochastic differential equations using physics-informed flow models
• Numerical methods for fluid dynamics. With applications to geophysics.
• On the eigenvector bias of Fourier feature networks: from regression to solving multi-scale PDEs with physics-informed neural networks
• PhyCRNet: physics-informed convolutional-recurrent network for solving spatiotemporal PDEs
• Physics-informed generative adversarial networks for stochastic differential equations
• Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
• Physics-informed variational inference for uncertainty quantification of stochastic differential equations
• PI-VAE: physics-informed variational auto-encoder for stochastic differential equations
• PPINN: parareal physics-informed neural network for time-dependent PDEs
• Quantifying total uncertainty in physics-informed neural networks for solving forward and inverse stochastic problems
• Simulator-free solution of high-dimensional stochastic elliptic partial differential equations using deep neural networks
• Understanding and Mitigating Gradient Flow Pathologies in Physics-Informed Neural Networks
• Understanding and mitigating gradient pathologies in physics-informed neural networks
• When and why PINNs fail to train: a neural tangent kernel perspective

Cited in

(9)

• Pre-training strategy for solving evolution equations based on physics-informed neural networks
• Exact enforcement of temporal continuity in sequential physics-informed neural networks
• The novel graph transformer-based surrogate model for learning physical systems
• Time integration schemes based on neural networks for solving partial differential equations on coarse grids
• Integral regularization PINNs for evolution equations
• RTC-fPINNs: respecting temporal causality fPINNs method for time fractional fourth order partial differential equations
• Solving time-dependent partial differential equations via hard-constrained physics-informed neural networks with adaptive causal weighting
• Preserving conservation laws in the time-evolving natural Galerkin method via relaxation and projection
• CI-PINN: Causal Integral Physics-Informed Neural Network for Solving Evolution Equations

This page was built for publication: CEENs: causality-enforced evolutional networks for solving time-dependent partial differential equations

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6557798)