Robust training of machine learning interatomic potentials with dimensionality reduction and stratified sampling

Qi, Ji; Ko, Tsz Wai; Wood, Brandon C.; Pham, Tuan Anh; Ong, Shyue Ping

doi:10.1038/s41524-024-01227-4

Robust training of machine learning interatomic potentials with dimensionality reduction and stratified sampling

Journal Article · Mon Feb 26 00:00:00 EST 2024 · npj Computational Materials

DOI:https://doi.org/10.1038/s41524-024-01227-4· OSTI ID:2311933

; ; Wood, Brandon C.; Pham, Tuan Anh;

Abstract

Machine learning interatomic potentials (MLIPs) enable accurate simulations of materials at scales beyond that accessible by ab initio methods and play an increasingly important role in the study and design of materials. However, MLIPs are only as accurate and robust as the data on which they are trained. Here, we present DImensionality-Reduced Encoded Clusters with sTratified (DIRECT) sampling as an approach to select a robust training set of structures from a large and complex configuration space. By applying DIRECT sampling on the Materials Project relaxation trajectories dataset with over one million structures and 89 elements, we develop an improved materials 3-body graph network (M3GNet) universal potential that extrapolates more reliably to unseen structures. We further show that molecular dynamics (MD) simulations with the M3GNet universal potential can be used instead of expensive ab initio MD to rapidly create a large configuration space for target systems. We combined this scheme with DIRECT sampling to develop a reliable moment tensor potential for titanium hydrides without the need for iterative augmentation of training structures. This work paves the way for robust high-throughput development of MLIPs across any compositional complexity.

View Journal Article

Sponsoring Organization:: USDOE

OSTI ID:: 2311933

Journal Information:: npj Computational Materials, Journal Name: npj Computational Materials Journal Issue: 1 Vol. 10; ISSN 2057-3960

Publisher:: Nature Publishing GroupCopyright Statement

Country of Publication:: United Kingdom

Language:: English

References (49)

Frustration in Super‐Ionic Conductors Unraveled by the Density of Atomistic States Wang, Shuo; Liu, Yunsheng; Mo, Yifei Angewandte Chemie, Vol. 135, Issue 15 https://doi.org/10.1002/ange.202215544	journal	February 2023
The H−Ti (Hydrogen-Titanium) system San-Martin, A.; Manchester, F. D. Bulletin of Alloy Phase Diagrams, Vol. 8, Issue 1 https://doi.org/10.1007/BF02868888	journal	February 1987
Efficiency of ab-initio total energy calculations for metals and semiconductors using a plane-wave basis set Kresse, G.; Furthmüller, J. Computational Materials Science, Vol. 6, Issue 1, p. 15-50 https://doi.org/10.1016/0927-0256(96)00008-0	journal	July 1996
Hydrogen and deuterium diffusion in titanium dihydrides/dideuterides Kaess, U.; Majer, G.; Stoll, M. Journal of Alloys and Compounds, Vol. 259, Issue 1-2 https://doi.org/10.1016/S0925-8388(97)00124-2	journal	August 1997
Perspectives on Titanium Science and Technology Banerjee, Dipankar; Williams, J. C. Acta Materialia, Vol. 61, Issue 3 https://doi.org/10.1016/j.actamat.2012.10.043	journal	February 2013
Active learning of linearly parametrized interatomic potentials Podryabinkin, Evgeny V.; Shapeev, Alexander V. Computational Materials Science, Vol. 140 https://doi.org/10.1016/j.commatsci.2017.08.031	journal	December 2017
Accelerating high-throughput searches for new alloys with active learning of interatomic potentials Gubaev, Konstantin; Podryabinkin, Evgeny V.; Hart, Gus L. W. Computational Materials Science, Vol. 156 https://doi.org/10.1016/j.commatsci.2018.09.031	journal	January 2019
Hydriding of titanium: Recent trends and perspectives in advanced characterization and multiscale modeling Zhu, Yakun; Wook Heo, Tae; Rodriguez, Jennifer N. Current Opinion in Solid State and Materials Science, Vol. 26, Issue 6 https://doi.org/10.1016/j.cossms.2022.101020	journal	December 2022
DeePMD-kit: A deep learning package for many-body potential energy representation and molecular dynamics Wang, Han; Zhang, Linfeng; Han, Jiequn Computer Physics Communications, Vol. 228 https://doi.org/10.1016/j.cpc.2018.03.016	journal	July 2018
LAMMPS - a flexible simulation tool for particle-based materials modeling at the atomic, meso, and continuum scales Thompson, Aidan P.; Aktulga, H. Metin; Berger, Richard Computer Physics Communications, Vol. 271 https://doi.org/10.1016/j.cpc.2021.108171	journal	February 2022
Spectral neighbor analysis method for automated generation of quantum-accurate interatomic potentials Thompson, A. P.; Swiler, L. P.; Trott, C. R. Journal of Computational Physics, Vol. 285 https://doi.org/10.1016/j.jcp.2014.12.018	journal	March 2015
Hydrogen diffusion in plutonium hydrides from first principles Tang, Jun; Qiu, Ruizhi; Chen, Jinfan Journal of Nuclear Materials, Vol. 557 https://doi.org/10.1016/j.jnucmat.2021.153247	journal	December 2021
Compositionally complex perovskite oxides: Discovering a new class of solid electrolytes with interface-enabled conductivity improvements Ko, Shu-Ting; Lee, Tom; Qi, Ji Matter, Vol. 6, Issue 7 https://doi.org/10.1016/j.matt.2023.05.035	journal	July 2023
Bridging the gap between simulated and experimental ionic conductivities in lithium superionic conductors Qi, J.; Banerjee, S.; Zuo, Y. Materials Today Physics, Vol. 21 https://doi.org/10.1016/j.mtphys.2021.100463	journal	November 2021
Graph Networks as a Universal Machine Learning Framework for Molecules and Crystals Chen, Chi; Ye, Weike; Zuo, Yunxing Chemistry of Materials, Vol. 31, Issue 9 https://doi.org/10.1021/acs.chemmater.9b01294	journal	April 2019
Machine Learning Force Fields Unke, Oliver T.; Chmiela, Stefan; Sauceda, Huziel E. Chemical Reviews https://doi.org/10.1021/acs.chemrev.0c01111	journal	March 2021
Performance and Cost Assessment of Machine Learning Interatomic Potentials Zuo, Yunxing; Chen, Chi; Li, Xiangguo The Journal of Physical Chemistry A, Vol. 124, Issue 4 https://doi.org/10.1021/acs.jpca.9b08723	journal	October 2019
Constructing High-Dimensional Neural Network Potential Energy Surfaces for Gas–Surface Scattering and Reactions Liu, Qinghua; Zhou, Xueyao; Zhou, Linsen The Journal of Physical Chemistry C, Vol. 122, Issue 3 https://doi.org/10.1021/acs.jpcc.7b12064	journal	January 2018
Thermodynamics and Kinetics of the Cathode–Electrolyte Interface in All-Solid-State Li–S Batteries Holekevi Chandrappa, Manas Likhit; Qi, Ji; Chen, Chi Journal of the American Chemical Society, Vol. 144, Issue 39 https://doi.org/10.1021/jacs.2c07482	journal	September 2022
E(3)-equivariant graph neural networks for data-efficient and accurate interatomic potentials Batzner, Simon; Musaelian, Albert; Sun, Lixin Nature Communications, Vol. 13, Issue 1 https://doi.org/10.1038/s41467-022-29939-5	journal	May 2022
Machine-learned interatomic potentials by active learning: amorphous and liquid hafnium dioxide Sivaraman, Ganesh; Krishnamoorthy, Anand Narayanan; Baur, Matthias npj Computational Materials, Vol. 6, Issue 1 https://doi.org/10.1038/s41524-020-00367-7	journal	July 2020
On-the-fly active learning of interpretable Bayesian force fields for atomistic rare events Vandermause, Jonathan; Torrisi, Steven B.; Batzner, Simon npj Computational Materials, Vol. 6, Issue 1 https://doi.org/10.1038/s41524-020-0283-z	journal	March 2020
Complex strengthening mechanisms in the NbMoTaW multi-principal element alloy Li, Xiang-Guo; Chen, Chi; Zheng, Hui npj Computational Materials, Vol. 6, Issue 1 https://doi.org/10.1038/s41524-020-0339-0	journal	June 2020
AtomSets as a hierarchical transfer learning framework for small and large materials datasets Chen, Chi; Ong, Shyue Ping npj Computational Materials, Vol. 7, Issue 1 https://doi.org/10.1038/s41524-021-00639-w	journal	October 2021
Training data selection for accuracy and transferability of interatomic potentials Montes de Oca Zapiain, David; Wood, Mitchell A.; Lubbers, Nicholas npj Computational Materials, Vol. 8, Issue 1 https://doi.org/10.1038/s41524-022-00872-x	journal	September 2022
Hyperactive learning for data-driven interatomic potentials van der Oord, Cas; Sachs, Matthias; Kovács, Dávid Péter npj Computational Materials, Vol. 9, Issue 1 https://doi.org/10.1038/s41524-023-01104-6	journal	September 2023
Origins of structural and electronic transitions in disordered silicon Deringer, Volker L.; Bernstein, Noam; Csányi, Gábor Nature, Vol. 589, Issue 7840 https://doi.org/10.1038/s41586-020-03072-z	journal	January 2021
CHGNet as a pretrained universal neural network potential for charge-informed atomistic modelling Deng, Bowen; Zhong, Peichen; Jun, KyuJung Nature Machine Intelligence, Vol. 5, Issue 9 https://doi.org/10.1038/s42256-023-00716-3	journal	September 2023
Learning properties of ordered and disordered materials from multi-fidelity data Chen, Chi; Zuo, Yunxing; Ye, Weike Nature Computational Science, Vol. 1, Issue 1 https://doi.org/10.1038/s43588-020-00002-x	journal	January 2021
A universal graph deep learning interatomic potential for the periodic table Chen, Chi; Ong, Shyue Ping Nature Computational Science, Vol. 2, Issue 11 https://doi.org/10.1038/s43588-022-00349-3	journal	November 2022
Uncertainty-driven dynamics for active learning of interatomic potentials Kulichenko, Maksim; Barros, Kipton; Lubbers, Nicholas Nature Computational Science, Vol. 3, Issue 3 https://doi.org/10.1038/s43588-023-00406-5	journal	March 2023
Diffusion of hydrogen in titanium, Ti88Al12 and Ti3Al Miyoshi, Toshiyuki; Naito, Shizuo; Yamamoto, Masahiro Journal of the Chemical Society, Faraday Transactions, Vol. 92, Issue 3 https://doi.org/10.1039/ft9969200483	journal	January 1996
Commentary: The Materials Project: A materials genome approach to accelerating materials innovation Jain, Anubhav; Ong, Shyue Ping; Hautier, Geoffroy APL Materials, Vol. 1, Issue 1 https://doi.org/10.1063/1.4812323	journal	July 2013
An analysis of hydrated proton diffusion in ab initio molecular dynamics Tse, Ying-Lung Steve; Knight, Chris; Voth, Gregory A. The Journal of Chemical Physics, Vol. 142, Issue 1 https://doi.org/10.1063/1.4905077	journal	January 2015
Reactive atomistic simulations of Diels-Alder reactions: The importance of molecular rotations Rivero, Uxía; Unke, Oliver T.; Meuwly, Markus The Journal of Chemical Physics, Vol. 151, Issue 10 https://doi.org/10.1063/1.5114981	journal	September 2019
An entropy-maximization approach to automated training set generation for interatomic potentials Karabin, Mariia; Perez, Danny The Journal of Chemical Physics, Vol. 153, Issue 9 https://doi.org/10.1063/5.0013059	journal	September 2020
Optimal data generation for machine learned interatomic potentials Allen, Connor; Bartók, Albert P. Machine Learning: Science and Technology, Vol. 3, Issue 4 https://doi.org/10.1088/2632-2153/ac9ae7	journal	December 2022
Hydrogen diffusion in bccTiHxandTi1−yVyHx Sevilla, E. H.; Cotts, R. M. Physical Review B, Vol. 37, Issue 12 https://doi.org/10.1103/PhysRevB.37.6813	journal	April 1988
Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set Kresse, G.; Furthmüller, J. Physical Review B, Vol. 54, Issue 16, p. 11169-11186 https://doi.org/10.1103/PhysRevB.54.11169	journal	October 1996
High-dimensional neural network potentials for metal surfaces: A prototype study for copper Artrith, Nongnuch; Behler, Jörg Physical Review B, Vol. 85, Issue 4 https://doi.org/10.1103/PhysRevB.85.045439	journal	January 2012
Lattice dynamics and electron-phonon coupling calculations using nondiagonal supercells Lloyd-Williams, Jonathan H.; Monserrat, Bartomeu Physical Review B, Vol. 92, Issue 18 https://doi.org/10.1103/PhysRevB.92.184301	journal	November 2015
Gaussian Approximation Potentials: The Accuracy of Quantum Mechanics, without the Electrons Bartók, Albert P.; Payne, Mike C.; Kondor, Risi Physical Review Letters, Vol. 104, Issue 13 https://doi.org/10.1103/PhysRevLett.104.136403	journal	April 2010
Generalized Gradient Approximation Made Simple Perdew, John P.; Burke, Kieron; Ernzerhof, Matthias Physical Review Letters, Vol. 77, Issue 18, p. 3865-3868 https://doi.org/10.1103/PhysRevLett.77.3865	journal	October 1996
Generalized Neural-Network Representation of High-Dimensional Potential-Energy Surfaces Behler, Jörg; Parrinello, Michele Physical Review Letters, Vol. 98, Issue 14 https://doi.org/10.1103/PhysRevLett.98.146401	journal	April 2007
Active learning of uniformly accurate interatomic potentials for materials simulation Zhang, Linfeng; Lin, De-Ye; Wang, Han Physical Review Materials, Vol. 3, Issue 2 https://doi.org/10.1103/PhysRevMaterials.3.023804	journal	February 2019
Moment Tensor Potentials: A Class of Systematically Improvable Interatomic Potentials Shapeev, Alexander V. Multiscale Modeling & Simulation, Vol. 14, Issue 3 https://doi.org/10.1137/15M1054183	journal	January 2016
BIRCH: an efficient data clustering method for very large databases Zhang, Tian; Ramakrishnan, Raghu; Livny, Miron ACM SIGMOD Record, Vol. 25, Issue 2 https://doi.org/10.1145/235968.233324	journal	June 1996
Solubility and Diffusion of Hydrogen in Pure Metals and Alloys Wipf, H. Physica Scripta, Vol. T94, Issue 1 https://doi.org/10.1238/Physica.Topical.094a00043	journal	January 2001
Mpf.2021.2.8 Chen, Chi; Ong, Shyue Ping figshare https://doi.org/10.6084/m9.figshare.19470599.v3	dataset	January 2022

Similar Records

Benchmarking structural evolution methods for training of machine learned interatomic potentials

Journal Article · Thu Jul 21 20:00:00 EDT 2022 · Journal of Physics. Condensed Matter · OSTI ID:1982442

Systematic softening in universal machine learning interatomic potentials

Journal Article · Thu Jan 09 19:00:00 EST 2025 · npj Computational Materials · OSTI ID:2528028

Robust training of machine learning interatomic potentials with dimensionality reduction and stratified sampling

Citation Formats

References (49)

Similar Records

Related Subjects