Extensible Structure-Informed Prediction of Formation Energy with Improved Accuracy and Usability employing Neural Networks

TL;DR

This paper introduces SIPFENN, a neural network tool that predicts formation energies of atomic structures with high accuracy, usability, and low computational cost, aiding materials discovery and analysis.

Contribution

The paper presents a new neural network-based model for formation energy prediction, optimized for accuracy, discovery, and efficiency, with an open-source implementation and user-friendly interface.

Findings

Achieved MAE of 28-42 meV/atom on OQMD data

Developed three models for different performance priorities

Reduced computational cost by a factor of 8

Abstract

In the present paper, we introduce a new neural network-based tool for the prediction of formation energies of atomic structures based on elemental and structural features of Voronoi-tessellated materials. We provide a concise overview of the connection between the machine learning and the true material-property relationship, how to improve the generalization accuracy by reducing overfitting, how new data can be incorporated into the model to tune it to a specific material system, and preliminary results on using models to preform local structure relaxations. The present work resulted in three final models optimized for (1) highest test accuracy on the Open Quantum Materials Database (OQMD), (2) performance in the discovery of new materials, and (3) performance at a low computational cost. On a test set of 21,800 compounds randomly selected from OQMD, they achieve a mean absolute error (MAE) of 28, 40, and 42 meV/atom, respectively. The second model provides better predictions in a test case of interest not present in the OQMD, while the third reduces the computational cost by a factor of 8. We collect our results in a new open-source tool called SIPFENN (Structure-Informed Prediction of Formation Energy using Neural Networks). SIPFENN not only improves the accuracy beyond existing models but also ships in a ready-to-use form with pre-trained neural networks and a GUI interface. By virtue of this, it can be included in DFT calculations routines at nearly no cost.