Practical density functional theory (DFT) owes its success to the groundbreaking work of Kohn and Sham that introduced the exact calculation of the non-interacting kinetic energy of the electrons using an auxiliary mean-field system. However, the full power of DFT will not be unleashed until the exact relationship between the electron density and the non-interacting kinetic energy is found. Various attempts have been made to approximate this functional, similar to the exchange–correlation functional, with much less success due to the larger contribution of kinetic energy and its more non-local nature. In this work, we propose a new and efficient regularization method to train density functionals based on deep neural networks, with particular interest in the kinetic-energy functional. The method is tested on (effectively) one-dimensional systems, including the hydrogen chain, non-interacting electrons, and atoms of the first two periods, with excellent results. For atomic systems, the generalizability of the regularization method is demonstrated by training also an exchange–correlation functional, and the contrasting nature of the two functionals is discussed from a machine-learning perspective.
REFERENCES
1.
N.
Mardirossian
and M.
Head-Gordon
, “Thirty years of density functional theory in computational chemistry: An overview and extensive assessment of 200 density functionals
,” Mol. Phys.
115
, 2315
–2372
(2017
).2.
C. J.
Cramer
and D. G.
Truhlar
, “Density functional theory for transition metals and transition metal chemistry
,” Phys. Chem. Chem. Phys.
11
, 10757
(2009
).3.
A.
Jain
, Y.
Shin
, and K. A.
Persson
, “Computational predictions of energy materials using density functional theory
,” Nat. Rev. Mater.
1
, 15004
(2016
).4.
P.
Hohenberg
and W.
Kohn
, “Inhomogeneous electron gas
,” Phys. Rev.
136
, B864
–B871
(1964
).5.
W.
Kohn
and L. J.
Sham
, “Self-consistent equations including exchange and correlation effects
,” Phys. Rev.
140
, A1133
–A1138
(1965
).6.
Y. A.
Wang
and E. A.
Carter
, “Orbital-free kinetic-energy density functional theory
,” in Theoretical Methods in Condensed Phase Chemistry
, Progress in Theoretical Chemistry and Physics Vol. 5
, edited by S. D.
Schwartz
(Springer
, 2002
), pp. 117
–184
.7.
S. B.
Sears
, R. G.
Parr
, and U.
Dinur
, “On the quantum-mechanical kinetic energy as a measure of the information in a distribution
,” Isr. J. Chem.
19
, 165
–173
(1980
).8.
B. M.
Deb
and S. K.
Ghosh
, “New method for the direct calculation of electron density in many-electron systems. I. Application to closed-shell atoms
,” Int. J. Quantum Chem.
23
, 1
–26
(1983
).9.
N.
March
, “The local potential determining the square root of the ground-state electron density of atoms and molecules from the Schrödinger equation
,” Phys. Lett. A
113
, 476
–478
(1986
).10.
M.
Levy
and H.
Ou-Yang
, “Exact properties of the Pauli potential for the square root of the electron density and the kinetic energy functional
,” Phys. Rev. A
38
, 625
–629
(1988
).11.
J. P.
Perdew
and L. A.
Constantin
, “Laplacian-level density functionals for the kinetic energy density and exchange-correlation energy
,” Phys. Rev. B
75
, 155109
(2007
).12.
H. I.
Francisco
, J.
Carmona-Espíndola
, and J. L.
Gázquez
, “Analysis of the kinetic energy functional in the generalized gradient approximation
,” J. Chem. Phys.
154
, 084107
(2021
).13.
E.
Chacón
, J. E.
Alvarellos
, and P.
Tarazona
, “Nonlocal kinetic energy functional for nonhomogeneous electron systems
,” Phys. Rev. B
32
, 7868
–7877
(1985
).14.
P.
García-González
, J. E.
Alvarellos
, and E.
Chacón
, “Nonlocal kinetic-energy-density functionals
,” Phys. Rev. B
53
, 9509
–9512
(1996
).15.
D.
García-Aldea
and J. E.
Alvarellos
, “Fully nonlocal kinetic energy density functionals: A proposal and a general assessment for atomic systems
,” J. Chem. Phys.
129
, 074103
(2008
).16.
F.
Sarcinella
, E.
Fabiano
, L. A.
Constantin
, and F.
Della Sala
, “Nonlocal kinetic energy functionals in real space using a Yukawa-potential kernel: Properties, linear response, and model functionals
,” Phys. Rev. B
103
, 155127
(2021
).17.
J.
Jumper
et al, “Highly accurate protein structure prediction with AlphaFold
,” Nature
596
, 583
–589
(2021
).18.
S.
Ravuri
et al, “Skilful precipitation nowcasting using deep generative models of radar
,” Nature
597
, 672
–677
(2021
).19.
Z.
Lin
, H.
Akin
, R.
Rao
, B.
Hie
, Z.
Zhu
, W.
Lu
, N.
Smetanin
, R.
Verkuil
, O.
Kabeli
, Y.
Shmueli
, A.
dos Santos Costa
, M.
Fazel-Zarandi
, T.
Sercu
, S.
Candido
, and A.
Rives
, “Evolutionary-scale prediction of atomic-level protein structure with a language model
,” Science
379
, 1123
–1130
(2023
).20.
J.
Degrave
et al, “Magnetic control of tokamak plasmas through deep reinforcement learning
,” Nature
602
, 414
–419
(2022
).21.
K.
Choo
, A.
Mezzacapo
, and G.
Carleo
, “Fermionic neural-network states for ab-initio electronic structure
,” Nat. Commun.
11
, 2368
(2020
).22.
D.
Pfau
, J. S.
Spencer
, A. G. D. G.
Matthews
, and W. M. C.
Foulkes
, “Ab initio solution of the many-electron Schrödinger equation with deep neural networks
,” Phys. Rev. Res.
2
, 033429
(2020
).23.
J.
Hermann
, Z.
Schätzle
, and F.
Noé
, “Deep-neural-network solution of the electronic Schrödinger equation
,” Nat. Chem.
12
, 891
–897
(2020
).24.
S.
Dick
and M.
Fernandez-Serra
, “Machine learning accurate exchange and correlation functionals of the electronic density
,” Nat. Commun.
11
, 3509
(2020
).25.
J.
Kirkpatrick
, B.
McMorrow
, D. H. P.
Turban
, A. L.
Gaunt
, J. S.
Spencer
, A. G. D. G.
Matthews
, A.
Obika
, L.
Thiry
, M.
Fortunato
, D.
Pfau
, L. R.
Castellanos
, S.
Petersen
, A. W. R.
Nelson
, P.
Kohli
, P.
Mori-Sánchez
, D.
Hassabis
, and A. J.
Cohen
, “Pushing the frontiers of density functionals by solving the fractional electron problem
,” Science
374
, 1385
–1389
(2021
).26.
M.
Kasim
and S.
Vinko
, “Learning the exchange-correlation functional from nature with fully differentiable density functional theory
,” Phys. Rev. Lett.
127
, 126403
(2021
).27.
J.
Sun
, A.
Ruzsinszky
, and J. P.
Perdew
, “Strongly constrained and appropriately normed semilocal density functional
,” Phys. Rev. Lett.
115
, 036402
(2015
).28.
J. P.
Perdew
, “Jacob’s ladder of density functional approximations for the exchange-correlation energy
,” AIP Conf. Proc.
577
, 1
–20
(2001
).29.
M. G.
Medvedev
, I. S.
Bushmarinov
, J.
Sun
, J. P.
Perdew
, and K. A.
Lyssenko
, “Density functional theory is straying from the path toward the exact functional
,” Science
355
, 49
–52
(2017
).30.
J. C.
Snyder
, M.
Rupp
, K.
Hansen
, K.-R.
Müller
, and K.
Burke
, “Finding density functionals with machine learning
,” Phys. Rev. Lett.
108
, 253002
(2012
).31.
L.
Li
, S.
Hoyer
, R.
Pederson
, R.
Sun
, E. D.
Cubuk
, P.
Riley
, and K.
Burke
, “Kohn-Sham equations as regularizer: Building prior knowledge into machine-learned physics
,” Phys. Rev. Lett.
126
, 036401
(2021
).32.
T.
Helgaker
, H.
Larsen
, J.
Olsen
, and P.
Jørgensen
, “Direct optimization of the AO density matrix in Hartree–Fock and Kohn–Sham theories
,” Chem. Phys. Lett.
327
, 397
–403
(2000
).33.
H.
Larsen
, J.
Olsen
, P.
Jørgensen
, and T.
Helgaker
, “Direct optimization of the atomic-orbital density matrix using the conjugate-gradient method with a multilevel preconditioner
,” J. Chem. Phys.
115
, 9685
–9697
(2001
).34.
H.
Jiang
and W.
Yang
, “Conjugate-gradient optimization method for orbital-free density functional calculations
,” J. Chem. Phys.
121
, 2030
–2036
(2004
).35.
A.
Paszke
et al, “PyTorch: An imperative style, high-performance deep learning library
,” in Advances in Neural Information Processing Systems 32
(Curran Associates, Inc.
, 2019
), pp. 8024
–8035
.36.
J. C.
Snyder
, M.
Rupp
, K.
Hansen
, L.
Blooston
, K.-R.
Müller
, and K.
Burke
, “Orbital-free bond breaking via machine learning
,” J. Chem. Phys.
139
, 224104
(2013
).37.
J.
Seino
, R.
Kageyama
, M.
Fujinami
, Y.
Ikabata
, and H.
Nakai
, “Semi-local machine-learned kinetic energy density functional with third-order gradients of electron density
,” J. Chem. Phys.
148
, 241705
(2018
).38.
S. A.
Ghasemi
and T. D.
Kühne
, “Artificial neural networks for the kinetic energy functional of non-interacting fermions
,” J. Chem. Phys.
154
, 074107
(2021
).39.
B.
Kalita
, L.
Li
, R. J.
McCarty
, and K.
Burke
, “Learning to approximate density functionals
,” Acc. Chem. Res.
54
, 818
–826
(2021
).40.
F.
Imoto
, M.
Imada
, and A.
Oshiyama
, “Order-N orbital-free density-functional calculations with machine learning of functional derivatives for semiconductors and metals
,” Phys. Rev. Res.
3
, 033198
(2021
).41.
X. H.
Wu
, Z. X.
Ren
, and P. W.
Zhao
, “Nuclear energy density functionals from machine learning
,” Phys. Rev. C
105
, L031303
(2022
).42.
P.
Pulay
, “Improved SCF convergence acceleration
,” J. Comput. Chem.
3
, 556
–560
(1982
).43.
D.
Hendrycks
and K.
Gimpel
, “Bridging nonlinearities and stochastic regularizers with Gaussian error linear units
,” arXiv:1606.08415 (2016
).44.
T. E.
Baker
, E. M.
Stoudenmire
, L. O.
Wagner
, K.
Burke
, and S. R.
White
, “One-dimensional mimicking of electronic structure: The case for exponentials
,” Phys. Rev. B
91
, 235141
(2015
).45.
P.
Elliott
, A.
Cangi
, S.
Pittalis
, E. K. U.
Gross
, and K.
Burke
, “Almost exact exchange at almost no computational cost in electronic structure
,” Phys. Rev. A
92
, 022513
(2015
).46.
J.
Nocedal
, “Updating quasi-Newton matrices with limited storage
,” Math. Comput.
35
, 773
–782
(1980
).47.
48.
O.
Čertík
, Theoretical physics
, 2010
, https://www.theoretical-physics.com/dev/quantum/hf.html\#hartree-potential-in-spherical-symmetry; accessed 07 10, 2022.49.
Q.
Sun
, T. C.
Berkelbach
, N. S.
Blunt
, G. H.
Booth
, S.
Guo
, Z.
Li
, J.
Liu
, J. D.
McClain
, E. R.
Sayfutyarova
, S.
Sharma
, S.
Wouters
, and G. K.-L.
Chan
, “PySCF: The Python-based simulations of chemistry framework
,” Wiley Interdiscip. Rev.: Comput. Mol. Sci.
8
, e1340
(2018
).50.
Q.
Sun
et al, “Recent developments in the PySCF program package
,” J. Chem. Phys.
153
, 024109
(2020
).© 2023 Author(s). Published under an exclusive license by AIP Publishing.
2023
Author(s)
You do not currently have access to this content.
