English
 
Help Privacy Policy Disclaimer
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT

Released

Journal Article

Insightful classification of crystal structures using deep learning

MPS-Authors
/persons/resource/persons192341

Ziletti,  Angelo
Theory, Fritz Haber Institute, Max Planck Society;

/persons/resource/persons22064

Scheffler,  Matthias
Theory, Fritz Haber Institute, Max Planck Society;

/persons/resource/persons21549

Ghiringhelli,  Luca M.
Theory, Fritz Haber Institute, Max Planck Society;

External Resource
No external resources are shared
Fulltext (restricted access)
There are currently no full texts shared for your IP range.
Fulltext (public)

s41467-018-05169-6.pdf
(Publisher version), 2MB

Supplementary Material (public)
There is no public supplementary material available
Citation

Ziletti, A., Kumar, D., Scheffler, M., & Ghiringhelli, L. M. (2018). Insightful classification of crystal structures using deep learning. Nature Communications, 9: 2775. doi:10.1038/s41467-018-05169-6.


Cite as: https://hdl.handle.net/11858/00-001M-0000-002D-E97E-9
Abstract
Computational methods that automatically extract knowledge from data are critical for enabling data-driven materials science. A reliable identification of lattice symmetry is a crucial first step for materials characterization and analytics. Current methods require a user-specified threshold, and are unable to detect "average symmetries" for defective structures. Here, we propose a new machine-learning-based approach to automatically classify structures by crystal symmetry. First, we represent crystals by a diffraction image, and then construct a deep-learning neural-network model for classification. Our approach is able to correctly classify a dataset comprising more than 80,000 structures, including heavily defective ones. The internal operations of the neural network are unraveled through attentive response maps, demonstrating that it uses the same landmarks a materials scientist would use, although never explicitly instructed to do so. Our study paves the way for crystal-structure recognition in computational and experimental big-data materials science.