Crystal structure prediction accelerated by Bayesian optimization

被引：138

作者：

Yamashita, Tomoki ^{[1
,2
]}

Sato, Nobuya ^{[3
]}

Kino, Hiori ^{[1
,4
]}

Miyake, Takashi ^{[1
,3
,4
]}

Tsuda, Koji ^{[1
,5
,6
]}

Oguchi, Tamio ^{[1
,2
]}

机构：

[1] Natl Inst Mat Sci, Res & Serv Div Mat Data & Integrated Syst, Ctr Mat Res Informat Integrat, 1-2-1 Sengen, Tsukuba, Ibaraki 3050047, Japan

[2] Osaka Univ, Inst Sci & Ind Res, 8-1 Mihogaoka, Ibaraki, Osaka 5670047, Japan

[3] Natl Inst Adv Ind Sci & Technol, Res Ctr Computat Design Adv Funct Mat, 1-1-1 Umezono, Tsukuba, Ibaraki 3058568, Japan

[4] Natl Inst Mat Sci, Elements Strategy Initiat Ctr Magnet Mat, 1-2-1 Sengen, Tsukuba, Ibaraki 3050047, Japan

[5] Univ Tokyo, Dept Computat Biol & Med Sci, Grad Sch Frontier Sci, 5-1-5 Kashiwanoha, Kashiwa, Chiba 2778568, Japan

[6] RIKEN, Ctr Adv Intelligence Project, Chuo Ku, 1-4-1 Nihombashi, Tokyo 1030027, Japan

来源：

PHYSICAL REVIEW MATERIALS | 2018年 / 2卷 / 01期

基金：

日本科学技术振兴机构;

关键词：

GLOBAL OPTIMIZATION; ENERGY; CHEMISTRY;

D O I：

10.1103/PhysRevMaterials.2.013803

中图分类号：

T [工业技术];

学科分类号：

08 ;

摘要：

We propose a crystal structure prediction method based on Bayesian optimization. Our method is classified as a selection-type algorithm which is different from evolution-type algorithms such as an evolutionary algorithm and particle swarm optimization. Crystal structure prediction with Bayesian optimization can efficiently select the most stable structure from a large number of candidate structures with a lower number of searching trials using a machine learning technique. Crystal structure prediction using Bayesian optimization combined with random search is applied to known systems such as NaCl and Y2Co17 to discuss the efficiency of Bayesian optimization. These results demonstrate that Bayesian optimization can significantly reduce the number of searching trials required to find the global minimum structure by 30-40% in comparison with pure random search, which leads to much less computational cost.

引用

页数：6

共 38 条

[1] Crystal structure prediction using the minima hopping method [J].