Finding Density Functionals with Machine Learning

Machine learning is used to approximate density functionals. For the model problem of the kinetic energy of non-interacting fermions in 1d, mean absolute errors below 1 kcal/mol on test densities similar to the training set are reached with fewer than 100 training densities. A predictor identifies if a test density is within the interpolation region. Via principal component analysis, a projected functional derivative finds highly accurate self-consistent densities. Challenges for application of our method to real electronic structure problems are discussed.

References(19)

Figures(4)

[1]

P. Hohenberg
,
W. Kohn

- Phys.Rev.B 136 (1964) 864

[2]

W. Kohn
,
L.J. Sham

- Phys.Rev.A 140 (1965) 1133

[3]

P. Elliott
,
D. Lee
,
A. Cangi
,
K. Burke

- Phys.Rev.Lett. 100 (2008) 256406

[4]

P.J. Stephens
,
F.J. Devlin
,
C.F. Chabalowski
,
M.J. Frisch

- J.Phys.Chem. 98 (1994) 11623

[5]

Generalized Gradient Approximation Made Simple

- Phys.Rev.Lett. 77 (1996) 18, 3865
•
DOI:
- 10.1103/PhysRevLett.77.3865

[6]

J.P. Perdew
,
A. Ruzsinszky

- Int.J.Quant.Chem. 110 (2010) 2801

[7]

Density-functional exchange-energy approximation with correct asymptotic behavior

A.D. Becke
(
- Queen's U., Kingston
)

- Phys.Rev.A 38 (1988) 3098-3100
•
DOI:
- 10.1103/PhysRevA.38.3098

[8]

Development of the Colle-Salvetti correlation-energy formula into a functional of the electron density

- Phys.Rev.B 37 (1988) 785-789
•
DOI:
- 10.1103/PhysRevB.37.785

[9]

Theor. Chem. Accounts 120, 215

Y. Zhao
,
D. Truhlar

[10]

Trans. Neural Network 12, 181

K.-R. Müller
,
S. Mika
,
G. Rätsch
,
K. Tsuda
,
B. Schölkopf

[11]

in Reviews in Computational Chemistry, K. Lipkowitz and T. Cundari (eds.)

O. Ivanciuc

[12]

Phys. Rev. Lett. (to be published)

M. Rupp
,
A. Tkatchenko
,
K.-R. Müller
,
O.A. von Lilienfeld

[13]

in New Developments in Quantum Chemistry, J. Paz and A. Hernández (eds.)

V. Karasiev
,
R. Jones
,
S. Trickey
,
F. Harris

[14]

Density Functional Theory: An Approach to the Quantum Many-Body Problem

R.M. Dreizler
,
E.K.U. Gross

[15]

See e.g

E. Hairer
,
P. Nørsett
,
P. Syvert Paul
,
G. Wanner

[16]

The Elements of Statistical Learning. Data Mining, Inference, and Prediction, 2nd ed

T. Hastie
,
R. Tibshirani
,
J. Friedman

[17]

See Supplemental Material (appended at the end of this manuscript) for information necessary to construct the MLA functional and more detail on the PCA projections and self-consistent densities

[18]

D. Lee
,
L.A. Constantin
,
J.P. Perdew
,
K. Burke

- J.Chem.Phys. 130 (2009) 034107

[19]

Finding Density Functionals with Machine Learning

- Phys.Rev.Lett. 108 (2012) 25, 253002
•
e-Print:
- 1112.5441
•
DOI:
- 10.1103/PhysRevLett.108.253002