Big Data Meets Quantum Chemistry Approximations: The Δ-Machine Learning Approach
Big Data Meets Quantum Chemistry Approximations: The Δ-Machine Learning Approach
复制标题
DOI：
10.1021/acs.jctc.5b00099
复制
发表时间：
2015-05-01
影响因子：
5.5
通讯作者：
von Lilienfeld, O. Anatole
中科院分区：
文献类型：
--
作者：
Ramakrishnan, Raghunathan;Dral, Pavlo O.;von Lilienfeld, O. Anatole
对所有可能分子的虚拟空间进行化学精确和全面的研究受到量子化学计算成本的严重限制。我们引入了一种复合策略，将机器学习校正添加到计算成本低廉的近似传统量子方法中。训练后，对于比训练所用的分子集大得多的分子集，可以对焓、自由能、熵和电子相关能进行高度准确的预测。对于多达 16k 种 C7H10O2 异构体的热化学性质，我们提供了可以达到化学准确性的数值证据。我们还以 HartreeFock 的计算成本预测了后 Hartree-Fock 方法中的电子关联能量，并且我们建立了分子熵和电子关联之间的定性关系。我们的方法的可转移性得到了证明，使用半经验量子化学和机器学习模型，对 134k 有机分子中的 1% 和 10% 进行训练，以密度泛函理论的精度水平重现所有剩余分子的焓。
Chemically accurate and comprehensive studies of the virtual space of all possible molecules are severely limited by the computational cost of quantum chemistry. We introduce a composite strategy that adds machine learning corrections to computationally inexpensive approximate legacy quantum methods. After training, highly accurate predictions of enthalpies, free energies, entropies, and electron correlation energies are possible, for significantly larger molecular sets than used for training. For thermochemical properties of up to 16k isomers of C7H10O2 we present numerical evidence that chemical accuracy can be reached. We also predict electron correlation energy in post Hartree-Fock methods, at the computational cost of HartreeFock, and we establish a qualitative relationship between molecular entropy and electron correlation. The transferability of our approach is demonstrated, using semiempirical quantum chemistry and machine learning models trained on 1 and 10% of 134k organic molecules, to reproduce enthalpies of all remaining molecules at density functional theory level of accuracy.