Equivariant Message Passing Neural Network for Crystal Material Discovery

Astrid Klipfel, Olivier Peltre, Najwa Harrati, Yaël Fregier, Adlane Sayede +1 more
2/1/2023
cs.LG

No official code found. Best implementation: aklipf/pegnn (4 stars, last commit 2023-02-01).

Abstract

Automatic material discovery with desired properties is a fundamental challenge for material sciences. Considerable attention has recently been devoted to generating stable crystal structures. While existing work has shown impressive success on supervised tasks such as property prediction, the progress on unsupervised tasks such as material generation is still hampered by the limited extent to which the equivalent geometric representations of the same crystal are considered. To address this challenge, we propose EMPNN a periodic equivariant message-passing neural network that learns crystal lattice deformation in an unsupervised fashion. Our model equivalently acts on lattice according to the deformation action that must be performed, making it suitable for crystal generation, relaxation and optimisation. We present experimental evaluations that demonstrate the effectiveness of our approach.

View on arXivView PDF

Code Implementations(1)

aklipf/pegnnCommunity100%
40PythonApr 7, 20223 years ago

Similar Papers

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

82% match2025

Periodic Materials Generation using Text-Guided Joint Diffusion Model

82% match1 repo82025

StrainTensorNet: Predicting crystal structure elastic properties using SE(3)-equivariant graph neural networks

82% match2023

Similarity Equivariant Graph Neural Networks for Homogenization of Metamaterials

82% match1 repo92024

Transformer-Enhanced Variational Autoencoder for Crystal Structure Prediction

81% match2025

Optimized Crystallographic Graph Generation for Material Science

81% match1 repo42023

Cite this paper

@article{klipfel2023equivariant,
  title  = {Equivariant Message Passing Neural Network for Crystal Material Discovery},
  author = {Astrid Klipfel and Olivier Peltre and Najwa Harrati and Yaël Fregier and Adlane Sayede and Zied Bouraoui},
  year   = {2023},
  eprint = {2302.00485},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2302.00485v1}
}

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/2302.00485.svg)](https://codeofpaper.com/paper/2302.00485)

Discussion