Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Universal Machine Learning Interatomic Potentials are Ready for Phonons

Antoine Loew, Dewen Sun, Hai-Chen Wang, Silvana Botti, Miguel A. L. Marques

TL;DR

The results reveal that some models achieve high accuracy in predicting harmonic phonon properties, however, others still exhibit substantial inaccuracies, which highlights the importance of considering phonon-related properties in the development of universal machine learning interatomic potentials.

Abstract

There has been an ongoing race for the past several years to develop the best universal machinelearning interatomic potential. This progress has led to increasingly accurate models for predictingenergy, forces, and stresses, combining innovative architectures with big data. Here, we benchmarkthese models on their ability to predict harmonic phonon properties, which are critical for under-standing the vibrational and thermal behavior of materials. Using around 10 000 ab initio phononcalculations, we evaluate model performance across various phonon-related parameters to test theuniversal applicability of these models. The results reveal that some models achieve high accuracyin predicting harmonic phonon properties. However, others still exhibit substantial inaccuracies,even if they excel in the prediction of the energy and the forces for materials close to dynamicalequilibrium. These findings highlight the importance of considering phonon-related properties inthe development of universal machine learning interatomic potentials.

Universal Machine Learning Interatomic Potentials are Ready for Phonons

TL;DR

The results reveal that some models achieve high accuracy in predicting harmonic phonon properties, however, others still exhibit substantial inaccuracies, which highlights the importance of considering phonon-related properties in the development of universal machine learning interatomic potentials.

Abstract

There has been an ongoing race for the past several years to develop the best universal machinelearning interatomic potential. This progress has led to increasingly accurate models for predictingenergy, forces, and stresses, combining innovative architectures with big data. Here, we benchmarkthese models on their ability to predict harmonic phonon properties, which are critical for under-standing the vibrational and thermal behavior of materials. Using around 10 000 ab initio phononcalculations, we evaluate model performance across various phonon-related parameters to test theuniversal applicability of these models. The results reveal that some models achieve high accuracyin predicting harmonic phonon properties. However, others still exhibit substantial inaccuracies,even if they excel in the prediction of the energy and the forces for materials close to dynamicalequilibrium. These findings highlight the importance of considering phonon-related properties inthe development of universal machine learning interatomic potentials.

Paper Structure

This paper contains 13 sections, 5 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Results
  3. Dataset and Its Properties
  4. Relative Performances of uMLIPs
  5. Discussion
  6. Methods
  7. Ab initio Dataset
  8. uMLIP evaluation
  9. Data Availability
  10. Code Availability
  11. Acknowledgements
  12. Author Contributions
  13. Competing Interests

Figures (5)

  • Figure 1: Distribution of (a) number of different chemical elements per unit cell, (b) crystal systems, and (c) band gaps calculated with the PBE functional for all the materials in the dataset.
  • Figure 2: Periodic tables showing the frequency of the chemical elements in the structures from the dataset. Elements in gray are absent from the dataset.
  • Figure 3: Violin plot of the errors in the volume of the unit cell per atoms, relatively to the PBE reference data.
  • Figure 4: Violin plots of the errors in (a) the maximum phonon frequency, (b) the vibrational entropy, (c) the Helmholtz free energy, (d) the heat capacity, (e) the density of states and (f) the average of the sound velocity on the 3 accoustic branches, relatively to the PBE reference data.
  • Figure 5: Highest frequencies predicted for each structure for all models and from the original PBEsol MDR database.