Atomic-position independent descriptor for machine learning of material properties

被引:44
作者
Jain, Ankit [1 ]
Bligaard, Thomas [2 ]
机构
[1] Stanford Univ, Dept Chem Engn, Stanford, CA 94305 USA
[2] SLAC Natl Accelerator Lab, SUNCAT Ctr Interface Sci & Catalysis, 2575 Sand Hill Rd, Menlo Pk, CA 94025 USA
关键词
D O I
10.1103/PhysRevB.98.214112
中图分类号
T [工业技术];
学科分类号
08 ;
摘要
The high-throughput screening of periodic inorganic solids using machine learning methods requires atomic positions to encode structural and compositional details into appropriate material descriptors. These atomic positions are not available a priori for new materials, which severely limits exploration of novel materials. We overcome this limitation by using only crystallographic symmetry information in the structural description of materials. We show that for materials with identical structural symmetry, machine learning is trivial, and accuracies similar to that of density functional theory calculations can be achieved by using only atomic numbers in the material description. For machine learning of formation energies of bulk crystalline solids, this simple material descriptor is able to achieve prediction mean absolute errors of only 0.07 eV/at on a test dataset consisting of more than 85 000 diverse materials. This atomic-position independent material descriptor presents a new route of materials discovery wherein millions of materials can be screened by training a machine learning model over a drastically reduced subspace of materials.
引用
收藏
页数:7
相关论文
共 25 条
[1]  
Abadi M., 2016, TENSORFLOW LARGESCAL
[2]   Bilbao crystallographic server: I. Databases and crystallographic computing programs [J].
Aroyo, MI ;
Perez-Mato, JM ;
Capillas, C ;
Kroumova, E ;
Ivantchev, S ;
Madariaga, G ;
Kirov, A ;
Wondratschek, H .
ZEITSCHRIFT FUR KRISTALLOGRAPHIE, 2006, 221 (01) :15-27
[3]   Finding Unprecedentedly Low-Thermal-Conductivity Half-Heusler Semiconductors via High-Throughput Materials Modeling [J].
Carrete, Jesus ;
Li, Wu ;
Mingo, Natalio ;
Wang, Shidong ;
Curtarolo, Stefano .
PHYSICAL REVIEW X, 2014, 4 (01)
[4]   Oxide enthalpy of formation and band gap energy as accurate descriptors of oxygen vacancy formation energetics [J].
Deml, Ann M. ;
Stevanovic, Vladan ;
Muhich, Christopher L. ;
Musgrave, Charles B. ;
O'Hayre, Ryan .
ENERGY & ENVIRONMENTAL SCIENCE, 2014, 7 (06) :1996-2004
[5]  
Downs RT, 2003, AM MINERAL, V88, P247
[6]   Crystal structure representations for machine learning models of formation energies [J].
Faber, Felix ;
Lindmaa, Alexander ;
von Lilienfeld, O. Anatole ;
Armiento, Rickard .
INTERNATIONAL JOURNAL OF QUANTUM CHEMISTRY, 2015, 115 (16) :1094-1101
[7]  
Fedorov E. S., 1971, ACA MONOGRAPH, V7, P50
[8]   Accelerated Materials Design of Lithium Superionic Conductors Based on First-Principles Calculations and Machine Learning Algorithms [J].
Fujimura, Koji ;
Seko, Atsuto ;
Koyama, Yukinori ;
Kuwabara, Akihide ;
Kishida, Ippei ;
Shitara, Kazuki ;
Fisher, Craig A. J. ;
Moriwake, Hiroki ;
Tanaka, Isao .
ADVANCED ENERGY MATERIALS, 2013, 3 (08) :980-985
[9]   Big Data of Materials Science: Critical Role of the Descriptor [J].
Ghiringhelli, Luca M. ;
Vybiral, Jan ;
Levchenko, Sergey V. ;
Draxl, Claudia ;
Scheffler, Matthias .
PHYSICAL REVIEW LETTERS, 2015, 114 (10)
[10]   The optimal one dimensional periodic table: a modified Pettifor chemical scale from data mining [J].
Glawe, Henning ;
Sanna, Antonio ;
Gross, E. K. U. ;
Marques, Miguel A. L. .
NEW JOURNAL OF PHYSICS, 2016, 18