Evaluation of Thermochemical Machine Learning for Potential Energy Curves and Geometry Optimization
Evaluation of Thermochemical Machine Learning for Potential Energy Curves and Geometry Optimization
复制标题
DOI：
10.1021/acs.jpca.0c10147
复制
发表时间：
2021-02-25
影响因子：
2.9
通讯作者：
Hutchison, Geoffrey R.
中科院分区：
文献类型：
--
作者：
Folmsbee, Dakota L.;Koes, David R.;Hutchison, Geoffrey R.
虽然许多机器学习（ML）方法，特别是深度神经网络，已经针对密度泛函和量子化学能量和性质进行了训练，但这些方法中的绝大多数都集中在单点能量上。原则上，这种ML方法一旦经过训练，就可以提供与密度泛函和波函数方法相当的热化学精度，但速度与传统力场或近似半经验方法相当。到目前为止，大多数努力都集中在优化的平衡单点能量和性质。在这项工作中，我们评估了几种领先的ML方法在一系列的键势能曲线和扭转势的准确性。这些方法在现有的ANI-1训练集上训练，使用omega B 97 X/6- 31 G（d）在非平衡几何形状下的单点计算。我们发现，在一系列小分子中，有几种方法既提供了定性准确性（例如，正确的最小值，排斥和吸引键区域，非谐形状，和单个最小值）和在最小值附近的平均绝对百分比误差方面的定量精度。目前，ANI-2x、FCHL和一种新的基于libmoldgrid的卷积神经网络Colorful CNN表现出良好的性能。
While many machine learning (ML) methods, particularly deep neural networks, have been trained for density functional and quantum chemical energies and properties, the vast majority of these methods focus on single-point energies. In principle, such ML methods, once trained, offer thermochemical accuracy on par with density functional and wave function methods but at speeds comparable to traditional force fields or approximate semiempirical methods. So far, most efforts have focused on optimized equilibrium single-point energies and properties. In this work, we evaluate the accuracy of several leading ML methods across a range of bond potential energy curves and torsional potentials. The methods were trained on the existing ANI-1 training set, calculated using the omega B97X/6-31G(d) single points at nonequilibrium geometries. We find that across a range of small molecules, several methods offer both qualitative accuracy (e.g., correct minima, both repulsive and attractive bond regions, anharmonic shape, and single minima) and quantitative accuracy in terms of the mean absolute percent error near the minima. At the moment, ANI-2x, FCHL, and a new libmolgrid-based convolutional neural net, the Colorful CNN, show good performance.