Faculty of Physics , University of Vienna , Boltzmanngasse 5 , 1090 Vienna , Austria.
Universität Göttingen , Institut für Physikalische Chemie, Theoretische Chemie , Tammannstraße 6 , 37077 Göttingen , Germany.
J Chem Theory Comput. 2019 Mar 12;15(3):1827-1840. doi: 10.1021/acs.jctc.8b00770. Epub 2019 Feb 7.
Neural networks and other machine learning approaches have been successfully used to accurately represent atomic interaction potentials derived from computationally demanding electronic structure calculations. Due to their low computational cost, such representations open the possibility for large scale reactive molecular dynamics simulations of processes with bonding situations that cannot be described accurately with traditional empirical force fields. Here, we present a library of functions developed for the implementation of neural network potentials. Written in C++, this library incorporates several strategies resulting in a very high efficiency of neural network potential-energy and force evaluations. Based on this library, we have developed an implementation of the neural network potential within the molecular dynamics package LAMMPS and demonstrate its performance using liquid water as a test system.
神经网络和其他机器学习方法已成功用于准确表示从计算要求高的电子结构计算中得出的原子相互作用势能。由于其计算成本低，因此这些表示形式为使用传统经验力场无法准确描述的键合情况的大规模反应分子动力学模拟开辟了可能性。在这里，我们介绍了为实现神经网络势而开发的函数库。该库用 C++编写，结合了几种策略，从而使神经网络势的能量和力评估非常高效。在此库的基础上，我们在分子动力学包 LAMMPS 中开发了神经网络势的实现，并使用液态水作为测试系统来演示其性能。
