ICML 2021spotlight1237 citations
E(n) Equivariant Graph Neural Networks
Vı́ctor Garcia Satorras, Emiel Hoogeboom, Max Welling
Abstract
This paper introduces a new model to learn graph neural networks equivariant to rotations, translations, reflections and permutations called E(n)-Equivariant Graph Neural Networks (EGNNs). In contrast with existing methods, our work does not require computationally expensive higher-order representations in intermediate layers while it still achieves competitive or better performance. In addition, whereas existing methods are limited to equivariance on 3 dimensional spaces, our model is easily scaled to higher-dimensional spaces. We demonstrate the effectiveness of our method on dynamical systems modelling, representation learning in graph autoencoders and predicting molecular properties.
BibTeX
@InProceedings{pmlr-v139-satorras21a,
title = {E(n) Equivariant Graph Neural Networks},
author = {Satorras, V\'{\i}ctor Garcia and Hoogeboom, Emiel and Welling, Max},
booktitle = {Proceedings of the 38th International Conference on Machine Learning},
pages = {9323--9332},
year = {2021},
editor = {Meila, Marina and Zhang, Tong},
volume = {139},
series = {Proceedings of Machine Learning Research},
month = {18--24 Jul},
publisher = {PMLR},
pdf = {http://proceedings.mlr.press/v139/satorras21a/satorras21a.pdf},
url = {https://proceedings.mlr.press/v139/satorras21a.html},
abstract = {This paper introduces a new model to learn graph neural networks equivariant to rotations, translations, reflections and permutations called E(n)-Equivariant Graph Neural Networks (EGNNs). In contrast with existing methods, our work does not require computationally expensive higher-order representations in intermediate layers while it still achieves competitive or better performance. In addition, whereas existing methods are limited to equivariance on 3 dimensional spaces, our model is easily scaled to higher-dimensional spaces. We demonstrate the effectiveness of our method on dynamical systems modelling, representation learning in graph autoencoders and predicting molecular properties.}
}