E(n) Equivariant Graph Neural Networks

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.

References in corpus (20)

• Semi-Supervised Classification with Graph Convolutional Networks
• Neural Message Passing for Quantum Chemistry
• Spectral Networks and Locally Connected Networks on Graphs
• Quantum-Chemical Insights from Deep Tensor Neural Networks
• Variational Graph Auto-Encoders
• Searching for Activation Functions
• Group Equivariant Convolutional Networks
• Tensor field networks: Rotation- and translation-equivariant neural networks for 3D point clouds
• GraphRNN: Generating Realistic Graphs with Deep Auto-regressive Models
• SE(3)-Transformers: 3D Roto-Translation Equivariant Attention Networks
• Neural Relational Inference for Interacting Systems
• Deep Reinforcement Learning in a Handful of Trials using Probabilistic Dynamics Models
• Cormorant: Covariant Molecular Neural Networks
• Covariant Compositional Networks For Learning Graphs
• Generalizing Convolutional Neural Networks for Equivariance to Lie Groups on Arbitrary Continuous Data
• Equivariant Flows: Exact Likelihood Generative Learning for Symmetric Densities
• Relevance of Rotationally Equivariant Convolutions for Predicting Molecular Properties
• Equivariant Flows: sampling configurations for multi-body systems with symmetric energies
• Graph Normalizing Flows
• On the Equivalence between Positional Node Embeddings and Structural Graph Representations

Cited by in corpus (28)

• Geometric Deep Learning: Grids, Groups, Graphs, Geodesics, and Gauges
• Evaluation of the MACE Force Field Architecture: from Medicinal Chemistry to Materials Science
• Pre-training Molecular Graph Representation with 3D Geometry
• 3D Infomax improves GNNs for Molecular Property Prediction
• Machine-learning hidden symmetries
• Equivariant Graph Neural Networks for 3D Macromolecular Structure
• Rotation Invariant Graph Neural Networks using Spin Convolutions
• Learning Gradient Fields for Molecular Conformation Generation
• Ensuring thermodynamic consistency with invertible coarse-graining
• Independent SE(3)-Equivariant Models for End-to-End Rigid Protein Docking
• Large-scale graph representation learning with very deep GNNs and self-supervision
• Geometric and Physical Quantities Improve E(3) Equivariant Message Passing
• Frame Averaging for Invariant and Equivariant Network Design
• Reflection-Equivariant Diffusion for 3D Structure Determination from Isotopologue Rotational Spectra in Natural Abundance
• E(n) Equivariant Normalizing Flows
• Learning Equivariant Energy Based Models with Equivariant Stein Variational Gradient Descent
• Equivariant Learning of Stochastic Fields: Gaussian Processes and Steerable Conditional Neural Processes
• Implicit Neural Representations for Chemical Reaction Paths
• Unifying O(3) Equivariant Neural Networks Design with Tensor-Network Formalism
• Equivariant Deep Dynamical Model for Motion Prediction
• Beyond permutation equivariance in graph networks
• Learning Large-Time-Step Molecular Dynamics with Graph Neural Networks
• Data efficiency in graph networks through equivariance
• Rotation Equivariant 3D Hand Mesh Generation from a Single RGB Image
• Symmetry-driven graph neural networks
• Physics-Augmented Learning: A New Paradigm Beyond Physics-Informed Learning
• ChebLieNet: Invariant Spectral Graph NNs Turned Equivariant by Riemannian Geometry on Lie Groups
• Learning Atomic Multipoles: Prediction of the Electrostatic Potential with Equivariant Graph Neural Networks