Insightful classification of crystal structures using deep learning

TL;DR

This paper introduces a deep learning method that automatically classifies crystal structures by symmetry from diffraction images, effectively handling defective and noisy data, advancing materials characterization.

Contribution

It presents a novel machine learning approach that uses diffraction images and neural networks to classify crystal symmetry, surpassing traditional threshold-based methods.

Findings

Successfully classified over 100,000 simulated structures including defective ones

Neural network uses symmetry landmarks similar to materials scientists

Method robustly handles noisy and incomplete structural data

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 machine-learning-based approach to automatically classify structures by crystal symmetry. First, we represent crystals by calculating a diffraction image, then construct a deep-learning neural-network model for classification. Our approach is able to correctly classify a dataset comprising more than 100 000 simulated crystal 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 of - possibly noisy and incomplete - three-dimensional structural data in big-data materials science.