Deep Potential: A General Representation of a Many-Body Potential Energy Surface

被引：220

作者：

Han, Jiequn ^{[1
]}

Zhang, Linfeng ^{[1
]}

Car, Roberto ^{[1
,2
,3
]}

Weinan, E. ^{[1
,4
,5
,6
,7
]}

机构：

[1] Princeton Univ, Program Appl & Computat Math, Princeton, NJ 08544 USA

[2] Princeton Univ, Dept Phys, Dept Chem, Princeton, NJ 08544 USA

[3] Princeton Univ, Princeton Inst Sci & Technol Mat, Princeton, NJ 08544 USA

[4] Princeton Univ, Dept Math, Princeton, NJ 08544 USA

[5] Peking Univ, Ctr Data Sci, Beijing 100871, Peoples R China

[6] Peking Univ, Beijing Int Ctr Math Res, Beijing 100871, Peoples R China

[7] Beijing Inst Big Data Res, Beijing 100871, Peoples R China

来源：

COMMUNICATIONS IN COMPUTATIONAL PHYSICS | 2018年 / 23卷 / 03期

关键词：

Potential energy surface; deep learning; molecular simulation; EMBEDDED-ATOM-METHOD; MOLECULAR-DYNAMICS; FORCE-FIELD;

D O I：

10.4208/cicp.OA-2017-0213

中图分类号：

O4 [物理学];

学科分类号：

0702 ;

摘要：

We present a simple, yet general, deep neural network representation of the potential energy surface for atomic and molecular systems. It is "first-principle" based, in the sense that no ad hoc approximations or empirical fitting functions are required. When tested on a wide variety of examples, it reproduces the original model within chemical accuracy. This brings us one step closer to carrying outmolecular simulations with quantum mechanics accuracy at empirical potential computational cost.

引用

页码：629 / 639

页数：11

共 42 条

[21] Communication: Understanding molecular representations in machine learning: The role of uniqueness and target similarity [J].

Huang, Bing ;

von Lilienfeld, O. Anatole .

JOURNAL OF CHEMICAL PHYSICS, 2016, 145 (16)

[22]

Ioffe Sergey, 2015, P MACHINE LEARNING R, V37, P448, DOI [DOI 10.48550/ARXIV.1502.03167, DOI 10.5555/3015118.3045167]

[23] On the determination of molecular fields - II From the equation of state of a gas [J].

Jones, JE .

PROCEEDINGS OF THE ROYAL SOCIETY OF LONDON SERIES A-CONTAINING PAPERS OF A MATHEMATICAL AND PHYSICAL CHARACTER, 1924, 106 (738) :463-477

[24]

Kingma D. P., P 3 INT C LEARN REPR

[25] SELF-CONSISTENT EQUATIONS INCLUDING EXCHANGE AND CORRELATION EFFECTS [J].

KOHN, W ;

SHAM, LJ .

PHYSICAL REVIEW, 1965, 140 (4A) :1133-&

[26] Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set [J].

Kresse, G ;

Furthmuller, J .

PHYSICAL REVIEW B, 1996, 54 (16) :11169-11186

[27] ImageNet Classification with Deep Convolutional Neural Networks [J].

Krizhevsky, Alex ;

Sutskever, Ilya ;

Hinton, Geoffrey E. .

COMMUNICATIONS OF THE ACM, 2017, 60 (06) :84-90

[28] All-atom empirical potential for molecular modeling and dynamics studies of proteins [J].

MacKerell, AD ;

Bashford, D ;

Bellott, M ;

Dunbrack, RL ;

Evanseck, JD ;

Field, MJ ;

Fischer, S ;

Gao, J ;

Guo, H ;

Ha, S ;

Joseph-McCarthy, D ;

Kuchnir, L ;

Kuczera, K ;

Lau, FTK ;

Mattos, C ;

Michnick, S ;

Ngo, T ;

Nguyen, DT ;

Prodhom, B ;

Reiher, WE ;

Roux, B ;

Schlenkrich, M ;

Smith, JC ;

Stote, R ;

Straub, J ;

Watanabe, M ;

Wiórkiewicz-Kuczera, J ;

Yin, D ;

Karplus, M .

JOURNAL OF PHYSICAL CHEMISTRY B, 1998, 102 (18) :3586-3616

[29] How van der Waals interactions determine the unique properties of water [J].

Morawietz, Tobias ;

Singraber, Andreas ;

Dellago, Christoph ;

Behler, Joerg .

PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA, 2016, 113 (30) :8368-8373

[30] Many Molecular Properties from One Kernel in Chemical Space [J].

Ramakrishnan, Raghunathan ;

von Lilienfeld, O. Anatole .

CHIMIA, 2015, 69 (04) :182-186

← 1 2 3 4 5 →