Wang Tzu Yu, Neville Simon P, Schuurman Michael S
Department of Chemistry and Biomolecular Sciences, University of Ottawa, Ottawa, Ontario K1N 6N5, Canada.
National Research Council Canada, 100 Sussex Dr., Ottawa, Ontario K1A 0R6, Canada.
J Phys Chem Lett. 2023 Sep 7;14(35):7780-7786. doi: 10.1021/acs.jpclett.3c01649. Epub 2023 Aug 24.
The machine learning of potential energy surfaces (PESs) has undergone rapid progress in recent years. The vast majority of this work, however, has been focused on the learning of ground state PESs. To reliably extend machine learning protocols to excited state PESs, the occurrence of seams of conical intersections between adiabatic electronic states must be correctly accounted for. This introduces a serious problem, for at such points, the adiabatic potentials are not differentiable to any order, complicating the application of standard machine learning methods. We show that this issue may be overcome by instead learning the coordinate-dependent coefficients of the characteristic polynomial of a simple decomposition of the potential matrix. We demonstrate that, through this approach, quantitatively accurate machine learning models of seams of conical intersection may be constructed.
近年来，势能面（PESs）的机器学习取得了快速进展。然而，这项工作的绝大多数都集中在基态势能面的学习上。为了将机器学习协议可靠地扩展到激发态势能面，必须正确考虑绝热电子态之间锥形交叉缝的出现。这带来了一个严重的问题，因为在这些点上，绝热势在任何阶数上都不可微，这使得标准机器学习方法的应用变得复杂。我们表明，通过学习势矩阵简单分解的特征多项式的坐标相关系数，可以克服这个问题。我们证明，通过这种方法，可以构建出定量准确的锥形交叉缝机器学习模型。
