Discrete gradient flow approximations of high dimensional evolution partial differential equations via deep neural networks

DOI10.1016/J.CNSNS.2022.106893MaRDI QIDQ2108629zbMATH OpenOpenAlexWikidataFDO

Authors Emmanuil H. Georgoulis, Michail Loulakis, Asterios Tsiourvas

Publication date 20 December 2022

Published in Communications in Nonlinear Science and Numerical Simulation (Search for Journal in Brave)

Copyright license Creative Commons Attribution 4.0 International

Full work available at URL https://arxiv.org/abs/2206.00290

zbMATH Keywords

deep neural networksnumerical analysispartial differential equations

Mathematics Subject Classification ID

Probabilistic methods, stochastic differential equations (65Cxx) Numerical methods for partial differential equations, initial value and time-dependent initial-boundary value problems (65Mxx) Artificial intelligence (68Txx)

Cites work

• Deep learning
• Deep neural network approximation for high-dimensional elliptic PDEs with boundary conditions
• Deep Nitsche Method: Deep Ritz Method with Essential Boundary Conditions
• DGM: a deep learning algorithm for solving partial differential equations
• Gradient flows in metric spaces and in the space of probability measures
• Numerical approximations of stochastic differential equations with non-globally Lipschitz continuous coefficients
• On atomistic-to-continuum couplings without ghost forces in three dimensions
• PFNN: a penalty-free neural network method for solving a class of second-order boundary-value problems on complex geometries
• Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
• The deep Ritz method: a deep learning-based numerical algorithm for solving variational problems
• The Sinkhorn–Knopp Algorithm: Convergence and Applications
• The Variational Formulation of the Fokker--Planck Equation
• Über ein Variationsprinzip zur Lösung von Dirichlet-Problemen bei Verwendung von Teilräumen, die keinen Randbedingungen unterworfen sind. (On a variational principle for solving Dirichlet problems less boundary conditions using subspaces)

Cited in

(10)

• Derivative-informed projected neural networks for high-dimensional parametric maps governed by PDEs
• Deep neural networks based temporal-difference methods for high-dimensional parabolic partial differential equations
• On the use of graph neural networks and shape‐function‐based gradient computation in the deep energy method
• Discrete Gradient Flow Approximations of High Dimensional Evolution Partial Differential Equations via Deep Neural Networks
• A time-stepping deep gradient flow method for option pricing in (rough) diffusion models
• Space-time marching-based neural direct method for the high-dimensional Hamilton-Jacobi-Bellman equation
• Neural network Poisson-Boltzmann electrostatics for biomolecular interactions
• Runge-Kutta physics informed neural networks: formulation and analysis
• A deep Uzawa-Lagrange multiplier approach for boundary conditions in pinns and deep Ritz methods
• Deep Ritz--finite element methods: neural network methods trained with finite elements

Describes a project that uses

Uses Software

• DGM

This page was built for publication: Discrete gradient flow approximations of high dimensional evolution partial differential equations via deep neural networks

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2108629)