Bridging deep learning force fields and electronic structures with a physics-informed approach

Machine Learning


  • Kohn, W., Becke, A. D. & Parr, R. G. Density functional theory of electronic structure. J. Phys. Chem. 100, 12974–12980 (1996).

    CAS 

    Google Scholar 

  • Neugebauer, J. & Hickel, T. Density functional theory in materials science. Wiley Interdiscip. Rev. Comput. Mol. Sci. 3, 438–448 (2013).

    CAS 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Carleo, G. et al. Machine learning and the physical sciences. Rev. Mod. Phys. 91, 045002 (2019).

    CAS 

    Google Scholar 

  • Schmidt, J., Marques, M. R., Botti, S. & Marques, M. A. Recent advances and applications of machine learning in solid-state materials science. Npj Comput. Mater. 5, 83 (2019).

    Google Scholar 

  • Behler, J. Four generations of high-dimensional neural network potentials. Chem. Rev. 121, 10037–10072 (2021).

    CAS 
    PubMed 

    Google Scholar 

  • Unke, O. T. et al. Machine learning force fields. Chem. Rev. 121, 10142–10186 (2021).

    CAS 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Zhang, L., Han, J., Wang, H., Car, R. & Weinan, E. Deep potential molecular dynamics: a scalable model with the accuracy of quantum mechanics. Phys. Rev. Lett. 120, 143001 (2018).

    CAS 
    PubMed 

    Google Scholar 

  • Artrith, N., Morawietz, T. & Behler, J. High-dimensional neural-network potentials for multicomponent systems: Applications to zinc oxide. Phys. Rev. B 83, 153101 (2011).

    Google Scholar 

  • Sosso, G. C., Miceli, G., Caravati, S., Behler, J. & Bernasconi, M. Neural network interatomic potential for the phase change material gete. Phys. Rev. B 85, 174103 (2012).

    Google Scholar 

  • Deringer, V. L., Caro, M. A. & Csányi, G. Machine learning interatomic potentials as emerging tools for materials science. Adv. Mater. 31, 1902765 (2019).

    CAS 

    Google Scholar 

  • Chen, C. & Ong, S. P. A universal graph deep learning interatomic potential for the periodic table. Nat. Comput. Sci. 2, 718–728 (2022).

    PubMed 

    Google Scholar 

  • Shapeev, A. V. Moment tensor potentials: A class of systematically improvable interatomic potentials. Multiscale Model. Simul. 14, 1153–1173 (2016).

    Google Scholar 

  • Thompson, A. P., Swiler, L. P., Trott, C. R., Foiles, S. M. & Tucker, G. J. Spectral neighbor analysis method for automated generation of quantum-accurate interatomic potentials. J. Comput. Phys. 285, 316–330 (2015).

    CAS 

    Google Scholar 

  • Behler, J. & Parrinello, M. Generalized neural-network representation of high-dimensional potential-energy surfaces. Phys. Rev. Lett. 98, 146401 (2007).

    PubMed 

    Google Scholar 

  • Bartók, A. P., Payne, M. C., Kondor, R. & Csányi, G. Gaussian approximation potentials: The accuracy of quantum mechanics, without the electrons. Phys. Rev. Lett. 104, 136403 (2010).

    PubMed 

    Google Scholar 

  • Unke, O. T. & Meuwly, M. Physnet: A neural network for predicting energies, forces, dipole moments, and partial charges. J. Chem. Theory Comput. 15, 3678–3693 (2019).

    CAS 
    PubMed 

    Google Scholar 

  • Schütt, K. T., Sauceda, H. E., Kindermans, P.-J., Tkatchenko, A. & Müller, K.-R. Schnet–a deep learning architecture for molecules and materials. J. Chem. Phys. 148, 241722 (2018).

    PubMed 

    Google Scholar 

  • Choudhary, K. et al. Unified graph neural network force-field for the periodic table: solid state applications. Digit. Discov. 2, 346–355 (2023).

    CAS 

    Google Scholar 

  • Park, C. W. et al. Accurate and scalable graph neural network force field and molecular dynamics with direct force architecture. Npj Comput. Mater. 7, 73 (2021).

    Google Scholar 

  • Haghighatlari, M. et al. Newtonnet: A newtonian message passing network for deep learning of interatomic potentials and forces. Digit. Discov. 1, 333–343 (2022).

    CAS 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Batzner, S. et al. E(3)-equivariant graph neural networks for data-efficient and accurate interatomic potentials. Nat. Commun. 13, 2453 (2022).

    CAS 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Husic, B. E. et al. Coarse graining molecular dynamics with graph neural networks. J. Chem. Phys. 153, 194101 (2020).

    CAS 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Xiao, H. et al. An invertible, invariant crystal representation for inverse design of solid-state materials using generative deep learning. Nat. Commun. 14, 7027 (2023).

    CAS 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Li, H. et al. Deep-learning density functional theory hamiltonian for efficient ab initio electronic-structure calculation. Nat. Comput. Sci. 2, 367–377 (2022).

    PubMed 
    PubMed Central 

    Google Scholar 

  • Li, H. et al. Deep-learning electronic-structure calculation of magnetic superstructures. Nat. Comput. Sci. 3, 321–327 (2023).

    PubMed 
    PubMed Central 

    Google Scholar 

  • Gong, X. et al. General framework for e(3)-equivariant neural network representation of density functional theory hamiltonian. Nat. Commun. 14, 2848 (2023).

    CAS 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Zhong, Y., Yu, H., Su, M., Gong, X. & Xiang, H. Transferable equivariant graph neural networks for the hamiltonians of molecules and solids. npj Comput. Mater. 9, 182 (2023).

    Google Scholar 

  • Prodan, E. & Kohn, W. Nearsightedness of electronic matter. Proc. Natl Acad. Sci. 102, 11635–11638 (2005).

    CAS 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Wang, H., Zhang, L., Han, J. & Weinan, E. Deepmd-kit: A deep learning package for many-body potential energy representation and molecular dynamics. Comput. Phys. Commun. 228, 178–184 (2018).

    CAS 

    Google Scholar 

  • Wu, J., Yang, J., Ma, L., Zhang, L. & Liu, S. Modular development of deep potential for complex solid solutions. Phys. Rev. B 107, 144102 (2023).

    CAS 

    Google Scholar 

  • Wu, J., Zhang, Y., Zhang, L. & Liu, S. Deep learning of accurate force field of ferroelectric hfo2. Phys. Rev. B 103, 024108 (2021).

    CAS 

    Google Scholar 

  • Xu, N., Shi, Y., He, Y. & Shao, Q. A deep-learning potential for crystalline and amorphous Li–Si alloys. J. Phys. Chem. C. 124, 16278–16288 (2020).

    CAS 

    Google Scholar 

  • Ouyang, X. et al. Quantum-accurate modeling of ferroelectric phase transition in perovskites from message-passing neural networks. J. Phys. Chem. C. 127, 20890–20902 (2023).

    CAS 

    Google Scholar 

  • Wang, Z. et al. Graph representation-based machine learning framework for predicting electronic band structures of quantum-confined nanostructures. Sci. China Mater. 65, 3157–3170 (2022).

    CAS 

    Google Scholar 

  • Knøsgaard, N. R. & Thygesen, K. S. Representing individual electronic states for machine learning GW band structures of 2d materials. Nat. Commun. 13, 468 (2022).

    PubMed 
    PubMed Central 

    Google Scholar 

  • Wang, Z. et al. Machine learning method for tight-binding hamiltonian parameterization from ab-initio band structure. npj Comput. Mater. 7, 11 (2021).

    CAS 

    Google Scholar 

  • Gong, W. et al. Graph transformer networks for accurate band structure prediction: An end-to-end approach. arXiv preprint arXiv:2411.16483 (2024).

  • Fu, L., Wu, Y., Shang, H. & Yang, J. Transformer-based neural-network quantum state method for electronic band structures of real solids. J. Chem. Theory Comput. 20, 6218–6226 (2024).

    CAS 
    PubMed 

    Google Scholar 

  • Carr, S., Fang, S. & Kaxiras, E. Electronic-structure methods for twisted moiré layers. Nat. Rev. Mater. 5, 748–763 (2020).

    CAS 

    Google Scholar 

  • Andrei, E. Y. & MacDonald, A. H. Graphene bilayers with a twist. Nat. Mater. 19, 1265–1275 (2020).

    CAS 
    PubMed 

    Google Scholar 

  • Bistritzer, R. & MacDonald, A. H. Moiré bands in twisted double-layer graphene. Proc. Natl Acad. Sci. 108, 12233–12237 (2011).

    CAS 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Kohn, W. Analytic properties of Bloch waves and Wannier functions. Phys. Rev. 115, 809 (1959).

    Google Scholar 

  • Souza, I., Marzari, N. & Vanderbilt, D. Maximally localized Wannier functions for entangled energy bands. Phys. Rev. B 65, 035109 (2001).

    Google Scholar 

  • Marzari, N., Mostofi, A. A., Yates, J. R., Souza, I. & Vanderbilt, D. Maximally localized wannier functions: Theory and applications. Rev. Mod. Phys. 84, 1419 (2012).

    CAS 

    Google Scholar 

  • Wang, H., Ma, S., Zhang, S. & Lei, D. Intrinsic superflat bands in general twisted bilayer systems. Light Sci. Appl. 11, 159 (2022).

    CAS 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Karniadakis, G. E. et al. Physics-informed machine learning. Nat. Rev. Phys. 3, 422–440 (2021).

    Google Scholar 

  • Wang, X., Yates, J. R., Souza, I. & Vanderbilt, D. Ab initio calculation of the anomalous hall conductivity by wannier interpolation. Phys. Rev. B 74, 195118 (2006).

    Google Scholar 

  • Zhou, J., Qiao, J., Bournel, A. & Zhao, W. Intrinsic spin hall conductivity of the semimetals MoTe2 and WTe2. Phys. Rev. B 99, 060408 (2019).

    CAS 

    Google Scholar 

  • Ibañez-Azpiroz, J., Tsirkin, S. S. & Souza, I. Ab initio calculation of the shift photocurrent by Wannier interpolation. Phys. Rev. B 97, 245143 (2018).

    Google Scholar 

  • Pizzi, G., Volja, D., Kozinsky, B., Fornari, M. & Marzari, N. BoltzWann: A code for the evaluation of thermoelectric and electronic transport properties with a maximally-localized Wannier functions basis. Comput. Phys. Commun. 185, 422–429 (2014).

    CAS 

    Google Scholar 

  • Giustino, F., Cohen, M. L. & Louie, S. G. Electron-phonon interaction using Wannier functions. Phys. Rev. B 76, 165108 (2007).

    Google Scholar 

  • Poncé, S., Margine, E. R., Verdi, C. & Giustino, F. Epw: Electron–phonon coupling, transport and superconducting properties using maximally localized wannier functions. Comput. Phys. Commun. 209, 116–133 (2016).

    Google Scholar 

  • Koshino, M. et al. Maximally localized Wannier orbitals and the extended Hubbard model for twisted bilayer graphene. Phys. Rev. X 8, 031087 (2018).

    CAS 

    Google Scholar 

  • Garrity, K. F. & Choudhary, K. Database of wannier tight-binding hamiltonians using high-throughput density functional theory. Sci. Data 8, 106 (2021).

    CAS 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Brouder, C., Panati, G., Calandra, M., Mourougane, C. & Marzari, N. Exponential localization of Wannier functions in insulators. Phys. Rev. Lett. 98, 046402 (2007).

    PubMed 

    Google Scholar 

  • Zhang, W., Yu, R., Zhang, H.-J., Dai, X. & Fang, Z. First-principles studies of the three-dimensional strong topological insulators Bi2Te3, Bi2Se3 and Sb2Te3. N. J. Phys. 12, 065013 (2010).

    Google Scholar 

  • Shi, W. et al. A charge-density-wave topological semimetal. Nat. Phys. 17, 381–387 (2021).

    CAS 

    Google Scholar 

  • Ponraj, J. S. et al. Photonics and optoelectronics of two-dimensional materials beyond graphene. Nanotechnol 27, 462001 (2016).

    Google Scholar 

  • Mostofi, A. A. et al. An updated version of wannier90: A tool for obtaining maximally-localised Wannier functions. Comput. Phys. Commun. 185, 2309–2310 (2014).

    CAS 

    Google Scholar 

  • Pizzi, G. et al. Wannier90 as a community code: new features and applications. J. Condens. Matter Phys. 32, 165902 (2020).

    CAS 

    Google Scholar 

  • Lechermann, F. et al. Dynamical mean-field theory using Wannier functions: A flexible route to electronic structure calculations of strongly correlated materials. Phys. Rev. B—Condens. Matter Mater. Phys. 74, 125120 (2006).

    Google Scholar 

  • Sporkmann, B. & Bross, H. Calculation of Wannier functions for fcc transition metals by Fourier transformation of Bloch functions. Phys. Rev. B 49, 10869 (1994).

    CAS 

    Google Scholar 

  • Sakuma, R. Symmetry-adapted Wannier functions in the maximal localization procedure. Phys. Rev. B 87, 235109 (2013).

    Google Scholar 

  • Lei, B.-H., Pan, S., Yang, Z., Cao, C. & Singh, D. J. Second harmonic generation susceptibilities from symmetry adapted wannier functions. Phys. Rev. Lett. 125, 187402 (2020).

    CAS 
    PubMed 

    Google Scholar 

  • Hinuma, Y., Grüneis, A., Kresse, G. & Oba, F. Band alignment of semiconductors from density-functional theory and many-body perturbation theory. Phys. Rev. B 90, 155405 (2014).

    Google Scholar 

  • Choe, D.-H., West, D. & Zhang, S. Band alignment and the built-in potential of solids. Phys. Rev. Lett. 121, 196802 (2018).

    CAS 
    PubMed 

    Google Scholar 

  • Zhang, J., Xie, W., Zhao, J. & Zhang, S. Band alignment of two-dimensional lateral heterostructures. 2D Mater. 4, 015038 (2016).

    Google Scholar 

  • Kim, D. S. et al. Electrostatic moiré potential from twisted hexagonal boron nitride layers. Nat. Mater. 23, 65–70 (2024).

    CAS 
    PubMed 

    Google Scholar 

  • Tilak, N., Li, G., Taniguchi, T., Watanabe, K. & Andrei, E. Y. Moiré potential, lattice relaxation, and layer polarization in marginally twisted MoS2 bilayers. Nano Lett. 23, 73–81 (2022).

    PubMed 

    Google Scholar 

  • Yang, M., Raucci, U. & Parrinello, M. Reactant-induced dynamics of lithium imide surfaces during the ammonia decomposition process. Nat. Catal. 6, 829–836 (2023).

    CAS 

    Google Scholar 

  • Niu, H., Bonati, L., Piaggi, P. M. & Parrinello, M. Ab initio phase diagram and nucleation of gallium. Nat. Commun. 11, 2654 (2020).

    CAS 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Rodriguez, A., Lam, S. & Hu, M. Thermodynamic and transport properties of lif and flibe molten salts with deep learning potentials. ACS Appl. Mater. Interfaces 13, 55367–55379 (2021).

    CAS 
    PubMed 

    Google Scholar 

  • Piaggi, P. M., Weis, J., Panagiotopoulos, A. Z., Debenedetti, P. G. & Car, R. Homogeneous ice nucleation in an ab initio machine-learning model of water. Proc. Natl Acad. Sci. 119, e2207294119 (2022).

    CAS 
    PubMed 
    PubMed Central 

    Google Scholar 

  • Chen, T. et al. Modeling the high-pressure solid and liquid phases of tin from deep potentials with ab initio accuracy. Phys. Rev. Mater. 7, 053603 (2023).

    CAS 

    Google Scholar 

  • Liu, R. et al. Structural and dynamic properties of solvated hydroxide and hydronium ions in water from ab initio modeling. J. Chem. Phys. 157, 024503 (2022).

    CAS 
    PubMed 

    Google Scholar 

  • Wu, J. et al. Accurate force field of two-dimensional ferroelectrics from deep learning. Phys. Rev. B 104, 174107 (2021).

    CAS 

    Google Scholar 

  • Liu, G.-B., Shan, W.-Y., Yao, Y., Yao, W. & Xiao, D. Three-band tight-binding model for monolayers of group-vib transition metal dichalcogenides. Phys. Rev. B 88, 085433 (2013).

    Google Scholar 

  • Lenosky, T. J. et al. Highly optimized tight-binding model of silicon. Phys. Rev. B 55, 1528 (1997).

    CAS 

    Google Scholar 

  • Tang, M. S., Wang, C. Z., Chan, C. T. & Ho, K. M. Environment-dependent tight-binding potential model. Phys. Rev. B 53, 979 (1996).

    CAS 

    Google Scholar 

  • Ribeiro, R. M. & Peres, N. M. Stability of boron nitride bilayers: Ground-state energies, interlayer distances, and tight-binding description. Phys. Rev. B 83, 235312 (2011).

    Google Scholar 

  • Hinton, G. E. & Salakhutdinov, R. R. Reducing the dimensionality of data with neural networks. science 313, 504–507 (2006).

    CAS 
    PubMed 

    Google Scholar 

  • Xie, T., Fu, X., Ganea, O.-E., Barzilay, R. & Jaakkola, T. Crystal diffusion variational autoencoder for periodic material generation. arXiv preprint arXiv:2110.06197 (2021).

  • Zhang, C., Bengio, S., Hardt, M., Recht, B. & Vinyals, O. Understanding deep learning (still) requires rethinking generalization. Commun. ACM 64, 107–115 (2021).

    Google Scholar 

  • Moradi, R., Berangi, R. & Minaei, B. A survey of regularization strategies for deep models. Artif. Intell. Rev. 53, 3947–3986 (2020).

    Google Scholar 

  • Vitale, V. et al. Automated high-throughput wannierisation. Npj Comput. Mater. 6, 66 (2020).

    Google Scholar 

  • Giannozzi, P. et al. Quantum espresso: A modular and open-source software project for quantum simulations of materials. J. Phys.: Condens. Matter 21, 395502–20 (2009).

    PubMed 

    Google Scholar 

  • Perdew, J. P., Burke, K. & Ernzerhof, M. Generalized gradient approximation made simple. Phys. Rev. Lett. 77, 3865 (1996).

    CAS 
    PubMed 

    Google Scholar 

  • Monkhorst, H. J. & Pack, J. D. Special points for Brillouin-zone integrations. Phys. Rev. B 13, 5188–5192 (1976).

    Google Scholar 

  • Grimme, S. Semiempirical GGA-type density functional constructed with a long-range dispersion correction. J. Comput. Chem. 27, 1787–1799 (2006).

    CAS 
    PubMed 

    Google Scholar 

  • Paszke, A. et al. Pytorch: An imperative style, high-performance deep learning library. arXiv preprint arXiv:1912.01703 (2019).



  • Source link

    Leave a Reply

    Your email address will not be published. Required fields are marked *