MOMENT TENSOR POTENTIALS: A CLASS OF SYSTEMATICALLY IMPROVABLE INTERATOMIC POTENTIALS

被引:1032
作者
Shapeev, Alexander V. [1 ]
机构
[1] Skolkovo Innovat Ctr, Skolkovo Inst Sci & Technol, Bldg 3, Moscow 143026, Russia
关键词
machine learning; interatomic potentials; moment tensor potentials;
D O I
10.1137/15M1054183
中图分类号
O1 [数学];
学科分类号
0701 ; 070101 ;
摘要
Density functional theory offers a very accurate way of computing materials properties from first principles. However, it is too expensive for modeling large-scale molecular systems whose properties are, in contrast, computed using interatomic potentials. The present paper considers, from a mathematical point of view, the problem of constructing interatomic potentials that approximate a given quantum-mechanical interaction model. In particular, a new class of systematically improvable potentials is proposed, analyzed, and tested on an existing quantum-mechanical database.
引用
收藏
页码:1153 / 1173
页数:21
相关论文
共 31 条
[1]   Slave mode expansion for obtaining ab initio interatomic potentials [J].
Ai, Xinyuan ;
Chen, Yue ;
Marianetti, Chris A. .
PHYSICAL REVIEW B, 2014, 90 (01)
[2]  
Alpaydin E, 2014, ADAPT COMPUT MACH LE, P1
[3]  
[Anonymous], 2008, Functions of matrices: theory and computation
[4]  
[Anonymous], 2015, ARXIV150202077
[5]   On representing chemical environments [J].
Bartok, Albert P. ;
Kondor, Risi ;
Csanyi, Gabor .
PHYSICAL REVIEW B, 2013, 87 (18)
[6]   Gaussian Approximation Potentials: The Accuracy of Quantum Mechanics, without the Electrons [J].
Bartok, Albert P. ;
Payne, Mike C. ;
Kondor, Risi ;
Csanyi, Gabor .
PHYSICAL REVIEW LETTERS, 2010, 104 (13)
[7]  
Bauer FL., 1960, NUMER MATH, V2, P137, DOI [10.1007/BF01386217, DOI 10.1007/BF01386217]
[8]   Representing potential energy surfaces by high-dimensional neural network potentials [J].
Behler, J. .
JOURNAL OF PHYSICS-CONDENSED MATTER, 2014, 26 (18)
[9]   Generalized neural-network representation of high-dimensional potential-energy surfaces [J].
Behler, Joerg ;
Parrinello, Michele .
PHYSICAL REVIEW LETTERS, 2007, 98 (14)
[10]   Atom-centered symmetry functions for constructing high-dimensional neural network potentials [J].
Behler, Joerg .
JOURNAL OF CHEMICAL PHYSICS, 2011, 134 (07)