English

Self-supervised learning for crystal property prediction via denoising

Machine Learning 2024-09-02 v1 Materials Science

Abstract

Accurate prediction of the properties of crystalline materials is crucial for targeted discovery, and this prediction is increasingly done with data-driven models. However, for many properties of interest, the number of materials for which a specific property has been determined is much smaller than the number of known materials. To overcome this disparity, we propose a novel self-supervised learning (SSL) strategy for material property prediction. Our approach, crystal denoising self-supervised learning (CDSSL), pretrains predictive models (e.g., graph networks) with a pretext task based on recovering valid material structures when given perturbed versions of these structures. We demonstrate that CDSSL models out-perform models trained without SSL, across material types, properties, and dataset sizes.

Keywords

Cite

@article{arxiv.2408.17255,
  title  = {Self-supervised learning for crystal property prediction via denoising},
  author = {Alexander New and Nam Q. Le and Michael J. Pekala and Christopher D. Stiles},
  journal= {arXiv preprint arXiv:2408.17255},
  year   = {2024}
}

Comments

Published at ICML 2024 AI4Science: https://openreview.net/forum?id=yML9ufAEoV

R2 v1 2026-06-28T18:28:47.375Z