Using a machine learning approach to determine the space group of a structure from the atomic pair distribution function. Issue 4 (26th June 2019)

• Record Type: Journal Article
• Authors: Liu, Chia-Hao; Tao, Yunzhe; Hsu, Daniel; Du, Qiang; Billinge, Simon J. L.
• Is Part Of: Acta crystallographica. Volume 75:Issue 4(2019:Jul.)
• Page Start: 633
• Page End: 643
• Publication Date: 2019-06-26
• Subjects: pair distribution function -- space groups -- convolutional neural network -- machine learning; Crystallography -- Periodicals; Condensed matter -- Periodicals; 548
• DOI: 10.1107/S2053273319005606
• Languages: English
• ISSNs: 2053-2733

Abstract

Abstract : We present applications of machine learning models for predicting the space group of the underlying structure from its atomic pair distribution function (PDF). Abstract : A method is presented for predicting the space group of a structure given a calculated or measured atomic pair distribution function (PDF) from that structure. The method utilizes machine learning models trained on more than 100 000 PDFs calculated from structures in the 45 most heavily represented space groups. In particular, a convolutional neural network (CNN) model is presented which yields a promising result in that it correctly identifies the space group among the top‐6 estimates 91.9% of the time. The CNN model also successfully identifies space groups for 12 out of 15 experimental PDFs. Interesting aspects of the failed estimates are discussed, which indicate that the CNN is failing in similar ways as conventional indexing algorithms applied to conventional powder diffraction data. This preliminary success of the CNN model shows the possibility of model‐independent assessment of PDF data on a wide class of materials.