Machine learning of molecular electronic properties in chemical compound space
Machine learning of molecular electronic properties in chemical compound space
复制标题
DOI：
10.1088/1367-2630/15/9/095003
复制
发表时间：
2013-09-04
影响因子：
3.3
通讯作者：
von Lilienfeld, O. Anatole
中科院分区：
文献类型：
--
作者：
Montavon, Gregoire;Rupp, Matthias;von Lilienfeld, O. Anatole
现代科学计算与电子结构理论的结合可以产生前所未有的数据量，可用于智能数据分析，以识别有意义的、新颖的和可预测的结构-性质关系。这样的关系使得能够在可合成的虚拟化合物池中高通量地筛选相关性质。在这里，我们提出了一个机器学习模型，该模型基于数千个有机分子的从头计算结果数据库进行训练，可以同时预测多个电子基态和激发态的性质。这些性质包括原子化能、极化率、前线轨道本征值、电离势、电子亲和势和激发能。机器学习模型基于深度多任务人工神经网络，利用各种分子属性之间的潜在相关性。输入与从头算方法相同，即所有原子的核电荷和笛卡尔坐标。对于有机小分子，这种“量子机器”的精确度与现代量子化学方法相似，有时甚至更高--而计算成本可以忽略不计。
The combination of modern scientific computing with electronic structure theory can lead to an unprecedented amount of data amenable to intelligent data analysis for the identification of meaningful, novel and predictive structure-property relationships. Such relationships enable high-throughput screening for relevant properties in an exponentially growing pool of virtual compounds that are synthetically accessible. Here, we present a machine learning model, trained on a database of ab initio calculation results for thousands of organic molecules, that simultaneously predicts multiple electronic ground- and excited-state properties. The properties include atomization energy, polarizability, frontier orbital eigenvalues, ionization potential, electron affinity and excitation energies. The machine learning model is based on a deep multi-task artificial neural network, exploiting the underlying correlations between various molecular properties. The input is identical to ab initio methods, i.e. nuclear charges and Cartesian coordinates of all atoms. For small organic molecules, the accuracy of such a 'quantum machine' is similar, and sometimes superior, to modern quantum-chemical methods-at negligible computational cost.