Machine Learning Interatomic Potential: Bridge the Gap Between Small-scale Models and Realistic Device-scale Simulations

Overview

Journal iScience

Publisher Cell Press

Date 2024 Apr 22

PMID 38646181

Authors

Guanjie Wang

Changrui Wang

Xuanguang Zhang

Zefeng Li

Jian Zhou

Zhimei Sun

Affiliations

Soon will be listed here.

Abstract

Machine learning interatomic potential (MLIP) overcomes the challenges of high computational costs in density-functional theory and the relatively low accuracy in classical large-scale molecular dynamics, facilitating more efficient and precise simulations in materials research and design. In this review, the current state of the four essential stages of MLIP is discussed, including data generation methods, material structure descriptors, six unique machine learning algorithms, and available software. Furthermore, the applications of MLIP in various fields are investigated, notably in phase-change memory materials, structure searching, material properties predicting, and the pre-trained universal models. Eventually, the future perspectives, consisting of standard datasets, transferability, generalization, and trade-off between accuracy and complexity in MLIPs, are reported.

Citing Articles

Is the Future of Materials Amorphous? Challenges and Opportunities in Simulations of Amorphous Materials.

Madanchi A, Azek E, Zongo K, Beland L, Mousseau N, Simine L ACS Phys Chem Au. 2025; 5(1):3-16.

PMID: 39867446 PMC: 11758375. DOI: 10.1021/acsphyschemau.4c00063.

References

Yao K, Herr J, Toth D, Mckintyre R, Parkhill J . The TensorMol-0.1 model chemistry: a neural network augmented with long-range physics. Chem Sci. 2018; 9(8):2261-2269. PMC: 5897848. DOI: 10.1039/c7sc04934j. View

Shields B, Stevens J, Li J, Parasram M, Damani F, Martinez Alvarado J . Bayesian reaction optimization as a tool for chemical synthesis. Nature. 2021; 590(7844):89-96. DOI: 10.1038/s41586-021-03213-y. View

Unke O, Meuwly M . PhysNet: A Neural Network for Predicting Energies, Forces, Dipole Moments, and Partial Charges. J Chem Theory Comput. 2019; 15(6):3678-3693. DOI: 10.1021/acs.jctc.9b00181. View

Pinheiro Jr M, Ge F, Ferre N, Dral P, Barbatti M . Choosing the right molecular machine learning potential. Chem Sci. 2021; 12(43):14396-14413. PMC: 8580106. DOI: 10.1039/d1sc03564a. View

Balyakin I, Rempel S, Ryltsev R, Rempel A . Deep machine learning interatomic potential for liquid silica. Phys Rev E. 2020; 102(5-1):052125. DOI: 10.1103/PhysRevE.102.052125. View

Maier W, Stowe K, Sieg S . Combinatorial and high-throughput materials science. Angew Chem Int Ed Engl. 2007; 46(32):6016-67. DOI: 10.1002/anie.200603675. View

Belsky A, Hellenbrandt M, Karen V, Luksch P . New developments in the Inorganic Crystal Structure Database (ICSD): accessibility in support of materials research and design. Acta Crystallogr B. 2002; 58(Pt 3 Pt 1):364-9. DOI: 10.1107/s0108768102006948. View

Liu Y, Yang J, Xin G, Liu L, Csanyi G, Cao B . Machine learning interatomic potential developed for molecular simulations on thermal properties of β-GaO. J Chem Phys. 2020; 153(14):144501. DOI: 10.1063/5.0027643. View

Qiao Z, Christensen A, Welborn M, Manby F, Anandkumar A, Miller 3rd T . Informing geometric deep learning with electronic interactions to accelerate quantum chemistry. Proc Natl Acad Sci U S A. 2022; 119(31):e2205221119. PMC: 9351474. DOI: 10.1073/pnas.2205221119. View

10.

Yao K, Herr J, Brown S, Parkhill J . Intrinsic Bond Energies from a Bonds-in-Molecules Neural Network. J Phys Chem Lett. 2017; 8(12):2689-2694. DOI: 10.1021/acs.jpclett.7b01072. View

11.

Tshitoyan V, Dagdelen J, Weston L, Dunn A, Rong Z, Kononova O . Unsupervised word embeddings capture latent knowledge from materials science literature. Nature. 2019; 571(7763):95-98. DOI: 10.1038/s41586-019-1335-8. View

12.

Ko T, Finkler J, Goedecker S, Behler J . General-Purpose Machine Learning Potentials Capturing Nonlocal Charge Transfer. Acc Chem Res. 2021; 54(4):808-817. DOI: 10.1021/acs.accounts.0c00689. View

13.

Quiros M, Grazulis S, Girdzijauskaite S, Merkys A, Vaitkus A . Using SMILES strings for the description of chemical connectivity in the Crystallography Open Database. J Cheminform. 2018; 10(1):23. PMC: 5959826. DOI: 10.1186/s13321-018-0279-6. View

14.

Larsen A, Mortensen J, Blomqvist J, Castelli I, Christensen R, Dulak M . The atomic simulation environment-a Python library for working with atoms. J Phys Condens Matter. 2017; 29(27):273002. DOI: 10.1088/1361-648X/aa680e. View

15.

Deng Y, Fu S, Guo J, Xu X, Li H . Anisotropic Collective Variables with Machine Learning Potential for Crystallization of Complex Ceramics. ACS Nano. 2023; 17(14):14099-14113. DOI: 10.1021/acsnano.3c04602. View

16.

Chen C, Zuo Y, Ye W, Li X, Ong S . Learning properties of ordered and disordered materials from multi-fidelity data. Nat Comput Sci. 2024; 1(1):46-53. DOI: 10.1038/s43588-020-00002-x. View

17.

Himanen L, Geurts A, Foster A, Rinke P . Data-Driven Materials Science: Status, Challenges, and Perspectives. Adv Sci (Weinh). 2019; 6(21):1900808. PMC: 6839624. DOI: 10.1002/advs.201900808. View

18.

Rupp M, Tkatchenko A, Muller K, Anatole von Lilienfeld O . Fast and accurate modeling of molecular atomization energies with machine learning. Phys Rev Lett. 2012; 108(5):058301. DOI: 10.1103/PhysRevLett.108.058301. View

19.

Zaverkin V, Kastner J . Gaussian Moments as Physically Inspired Molecular Descriptors for Accurate and Scalable Machine Learning Potentials. J Chem Theory Comput. 2020; 16(8):5410-5421. DOI: 10.1021/acs.jctc.0c00347. View

20.

Unke O, Chmiela S, Sauceda H, Gastegger M, Poltavsky I, Schutt K . Machine Learning Force Fields. Chem Rev. 2021; 121(16):10142-10186. PMC: 8391964. DOI: 10.1021/acs.chemrev.0c01111. View