wACSF-Weighted atom-centered symmetry functions as descriptors in machine learning potentials

被引:228
作者
Gastegger, M. [1 ]
Schwiedrzik, L. [1 ]
Bittermann, M. [1 ]
Berzsenyi, F. [1 ]
Marquetand, P. [1 ]
机构
[1] Univ Vienna, Inst Theoret Chem, Fac Chem, Wahringer Str 17, A-1090 Vienna, Austria
关键词
NEURAL-NETWORK POTENTIALS; ENERGY SURFACES; MODELS; CHEMISTRY; ACCURACY;
D O I
10.1063/1.5019667
中图分类号
O64 [物理化学(理论化学)、化学物理学];
学科分类号
070304 ; 081704 ;
摘要
We introduce weighted atom-centered symmetry functions (wACSFs) as descriptors of a chemical system's geometry for use in the prediction of chemical properties such as enthalpies or potential energies via machine learning. The wACSFs are based on conventional atom-centered symmetry functions (ACSFs) but overcome the undesirable scaling of the latter with an increasing number of different elements in a chemical system. The performance of these two descriptors is compared using them as inputs in high-dimensional neural network potentials (HDNNPs), employing the molecular structures and associated enthalpies of the 133 855 molecules containing up to five different elements reported in the QM9 database as reference data. A substantially smaller number of wACSFs than ACSFs is needed to obtain a comparable spatial resolution of the molecular structures. At the same time, this smaller set of wACSFs leads to a significantly better generalization performance in the machine learning potential than the large set of conventional ACSFs. Furthermore, we show that the intrinsic parameters of the descriptors can in principle be optimized with a genetic algorithm in a highly automated manner. For the wACSFs employed here, we find however that using a simple empirical parametrization scheme is sufficient in order to obtain HDNNPs with high accuracy. Published by AIP Publishing.
引用
收藏
页数:11
相关论文
共 56 条
  • [1] [Anonymous], ARXIV171000017STATML
  • [2] [Anonymous], 2016, ABS160502688 THEAN D
  • [3] [Anonymous], 1964, On the Origin of Species, DOI 10.5962/bhl.title.82303
  • [4] [Anonymous], 2017, ARXIV170310603CSLG
  • [5] [Anonymous], 2006, Pattern Recognition and Machine Learning
  • [6] [Anonymous], 2008, HDB MOL DESCRIPTORS
  • [7] Efficient and accurate machine-learning interpolation of atomic energies in compositions with many species
    Artrith, Nongnuch
    Urban, Alexander
    Ceder, Gerbrand
    [J]. PHYSICAL REVIEW B, 2017, 96 (01)
  • [8] Gaussian approximation potentials: A brief tutorial introduction
    Bartok, Albert P.
    Csanyi, Gabor
    [J]. INTERNATIONAL JOURNAL OF QUANTUM CHEMISTRY, 2015, 115 (16) : 1051 - 1057
  • [9] On representing chemical environments
    Bartok, Albert P.
    Kondor, Risi
    Csanyi, Gabor
    [J]. PHYSICAL REVIEW B, 2013, 87 (18)
  • [10] Representing potential energy surfaces by high-dimensional neural network potentials
    Behler, J.
    [J]. JOURNAL OF PHYSICS-CONDENSED MATTER, 2014, 26 (18)