# Using a machine learning approach to determine the space group of a structure from the atomic pair distribution function (PDF)

## TL;DR

This paper introduces a machine learning method, specifically a CNN, to predict the space group of a structure from its atomic pair distribution function, achieving high accuracy on both calculated and experimental data.

## Contribution

It presents a novel CNN-based approach trained on extensive PDF data to accurately identify space groups, demonstrating potential for model-independent material analysis.

## Key findings

- CNN correctly identifies space group among top-6 estimates 91.9% of the time
- Successfully predicts space groups for 12 out of 15 experimental PDFs
- Indicates CNN's failure modes are similar to conventional algorithms

## Abstract

We present a method 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, we present a convolutional neural network (CNN) model which yields a promising result 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 on 12 out of 15 experimental PDFs. We discuss interesting aspects of the failed estimates, 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.