Multi-Scale Message Passing Neural PDE Solvers

Abstract

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 incorporating multi-scale sequence models and graph gating modules in the encoder and processor, respectively. Benchmark numerical experiments are presented to demonstrate that the proposed algorithm outperforms baselines, particularly on a PDE with a range of spatial and temporal scales.

Cite

Text

Equer et al. "Multi-Scale Message Passing Neural PDE Solvers." ICLR 2023 Workshops: Physics4ML, 2023.

Markdown

[Equer et al. "Multi-Scale Message Passing Neural PDE Solvers." ICLR 2023 Workshops: Physics4ML, 2023.](https://mlanthology.org/iclrw/2023/equer2023iclrw-multiscale/)

BibTeX

@inproceedings{equer2023iclrw-multiscale,
  title     = {{Multi-Scale Message Passing Neural PDE Solvers}},
  author    = {Equer, Léonard and Rusch, T. Konstantin and Mishra, Siddhartha},
  booktitle = {ICLR 2023 Workshops: Physics4ML},
  year      = {2023},
  url       = {https://mlanthology.org/iclrw/2023/equer2023iclrw-multiscale/}
}