FAENet: Frame Averaging Equivariant GNN for Materials Modeling

Alexandre Duval, Victor Schmidt, Alex Hernandez Garcia, Santiago Miret, Fragkiskos D. Malliaros +2 more
4/28/2023
cs.LG

No official code found. Best implementation: RolnickLab/ocp (7 stars, MIT, last commit 2026-03-15).

Abstract

Applications of machine learning techniques for materials modeling typically involve functions known to be equivariant or invariant to specific symmetries. While graph neural networks (GNNs) have proven successful in such tasks, they enforce symmetries via the model architecture, which often reduces their expressivity, scalability and comprehensibility. In this paper, we introduce (1) a flexible framework relying on stochastic frame-averaging (SFA) to make any model E(3)-equivariant or invariant through data transformations. (2) FAENet: a simple, fast and expressive GNN, optimized for SFA, that processes geometric information without any symmetrypreserving design constraints. We prove the validity of our method theoretically and empirically demonstrate its superior accuracy and computational scalability in materials modeling on the OC20 dataset (S2EF, IS2RE) as well as common molecular modeling tasks (QM9, QM7-X). A package implementation is available at https://faenet.readthedocs.io.

View on arXivView PDF

Code Implementations(1)

RolnickLab/ocpCommunity100%
MIT

FAENet code and training, based on OCP.

73Shell, PythonMar 2, 20227 months agoMIT

Similar Papers

Similarity Equivariant Graph Neural Networks for Homogenization of Metamaterials

83% match1 repo92024

Improving Molecular Modeling with Geometric GNNs: an Empirical Study

82% match2024

Quantized SO(3)-Equivariant Graph Neural Networks for Efficient Molecular Property Prediction

82% match2026

Equivariant graph convolutional neural networks for the representation of homogenized anisotropic microstructural mechanical response

82% match2024

Facet: highly efficient E(3)-equivariant networks for interatomic potentials

81% match2025

Energy-conserving equivariant GNN for elasticity of lattice architected metamaterials

81% match1 repo92024

Cite this paper

@article{duval2023faenet,
  title  = {FAENet: Frame Averaging Equivariant GNN for Materials Modeling},
  author = {Alexandre Duval and Victor Schmidt and Alex Hernandez Garcia and Santiago Miret and Fragkiskos D. Malliaros and Yoshua Bengio and David Rolnick},
  year   = {2023},
  eprint = {2305.05577},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2305.05577v1},
  journal = {ICML 2023 2023}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2305.05577.svg)](https://codeofpaper.com/paper/2305.05577)

Discussion