Finding Density Functionals with Machine Learning

被引：470

作者：

Snyder, John C. ^{[1
,2
]}

Rupp, Matthias ^{[3
,4
]}

Hansen, Katja ^{[3
]}

Mueller, Klaus-Robert ^{[3
,5
]}

Burke, Kieron ^{[1
,2
]}

机构：

[1] Univ Calif Irvine, Dept Chem, Irvine, CA 92697 USA

[2] Univ Calif Irvine, Dept Phys, Irvine, CA 92697 USA

[3] Tech Univ Berlin, Machine Learning Grp, D-10587 Berlin, Germany

[4] ETH, Inst Pharmaceut Sci, CH-8093 Zurich, Switzerland

[5] Korea Univ, Dept Brain & Cognit Engn, Seoul 136713, South Korea

来源：

PHYSICAL REVIEW LETTERS | 2012年 / 108卷 / 25期

基金：

美国国家科学基金会;

关键词：

APPROXIMATION;

D O I：

10.1103/PhysRevLett.108.253002

中图分类号：

O4 [物理学];

学科分类号：

0702 ;

摘要：

Machine learning is used to approximate density functionals. For the model problem of the kinetic energy of noninteracting 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. The challenges for application of our method to real electronic structure problems are discussed.

引用

页数：5

共 22 条

[1] DENSITY-FUNCTIONAL EXCHANGE-ENERGY APPROXIMATION WITH CORRECT ASYMPTOTIC-BEHAVIOR [J].

BECKE, AD .

PHYSICAL REVIEW A, 1988, 38 (06) :3098-3100

[2]

Braun ML, 2008, J MACH LEARN RES, V9, P1875

[3] Perspective on density functional theory [J].

Burke, Kieron .

JOURNAL OF CHEMICAL PHYSICS, 2012, 136 (15)

[4]

Dreizler R., 1990, DENSITY FUNCTIONAL T

[5] Semiclassical origins of density functionals [J].

Elliott, Peter ;

Lee, Donghyung ;

Cangi, Attila ;

Burke, Kieron .

PHYSICAL REVIEW LETTERS, 2008, 100 (25)

[6]

Hairer Ernst, 2000, Solving Ordinary Differential Equations I, Nonstiff Problems

[7]

Hastie T., 2009, ELEMENTS STAT LEARNI, DOI DOI 10.1007/978-0-387-84858-7

[8] INHOMOGENEOUS ELECTRON-GAS [J].

RAJAGOPAL, AK ;

CALLAWAY, J .

PHYSICAL REVIEW B, 1973, 7 (05) :1912-1919

[9]

Ivanciuc O., 2007, REV COMPUTATIONAL CH, P291, DOI [10.1002/9780470116449.ch6, DOI 10.1002/9780470116449.CH6]

[10]

KARASIEV VV, NEW DEV QUA IN PRESS

← 1 2 3 →