paper-with-me

홈 › Papers

Learning time-dependent PDE solver using Message Passing Graph Neural Networks

2022-04-15 · Pourya Pilva, Ahmad Zareei

One of the main challenges in solving time-dependent partial differential equations is to develop computationally efficient solvers that are accurate and stable. Here, we introduce a graph neural network approach to finding efficient PDE solvers through learning using message-passing models. We first introduce domain invariant features for PDE-data inspired by classical PDE solvers for an efficient physical representation. Next, we use graphs to represent PDE-data on an unstructured mesh and show that message passing graph neural networks (MPGNN) can parameterize governing equations, and as a result, efficiently learn accurate solver schemes for linear/nonlinear PDEs. We further show that the solvers are independent of the initial trained geometry, i.e. the trained solver can find PDE solution on different complex domains. Lastly, we show that a recurrent graph neural network approach can find a temporal sequence of solutions to a PDE.

📄 PDF Abstract BibTeX arXiv:2204.07651

Code (0)

등록된 구현이 없습니다.

Tasks

Graph Neural Network

Methods 이 논문이 사용한 방법론

Graph Neural Network 설명 없음

Similar Papers 제목 키워드 기반

Multi-Scale Message Passing Neural PDE Solvers

2023-02-07 · Léonard Equer, T. Konstantin Rusch, Siddhartha Mishra

We propose a novel multi-scale message passing neural network algorithm for learning the solutions of time-dependent PDEs. Our algorithm possesses both temporal and spatial multi-scale resolution features by incorporatin…

Autoregressive Renaissance in Neural PDE Solvers

2023-10-30 · Yolanne Yi Ran Lee

Recent developments in the field of neural partial differential equation (PDE) solvers have placed a strong emphasis on neural operators. However, the paper "Message Passing Neural PDE Solver" by Brandstetter et al. publ…

Graph Neural Network

Framelet Message Passing

2023-02-28 · Xinliang Liu, Bingxin Zhou, Chutian Zhang, Yu Guang Wang

Graph neural networks (GNNs) have achieved champion in wide applications. Neural message passing is a typical key module for feature propagation by aggregating neighboring features. In this work, we propose a new message…

Node Classification

On the under-reaching phenomenon in message-passing neural PDE solvers: revisiting the CFL condition

2025-07-09 · Lucas Tesan, Mikel M. Iparraguirre, David Gonzalez, Pedro Martins 외 arxiv

This paper proposes sharp lower bounds for the number of message passing iterations required in graph neural networks (GNNs) when solving partial differential equations (PDE). This significantly reduces the need for exha…

Message Passing Neural PDE Solvers

2022-02-07 · ICLR 2022 4 · Johannes Brandstetter, Daniel Worrall, Max Welling

The numerical solution of partial differential equations (PDEs) is difficult, having led to a century of research so far. Recently, there have been pushes to build neural--numerical hybrid solvers, which piggy-backs the …

Domain Adaptation