E(n) Equivariant Graph Neural Networks

Code

• github.com/vgsatorras/egnn
  Officialpytorch★ 526
• github.com/lucidrains/egnn-pytorch
  pytorch★ 520
• github.com/gerkone/egnn-jax
  jax★ 14
• github.com/stdereka/egnn
  pytorch★ 3
• github.com/oumarkaba/channels_egnn
  pytorch★ 2

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.

Tasks

Benchmark Results

Reproductions