arXiv：cs.LG· Rongzhi Gao, Shuguang Chen, Yang Zhou, GuanHua Chen, Ziyang Hu, ChiYung Yam·· 4 小时前AI 评分30
PEACE：用宇称分辨哈密顿量协变学习非绝热流形
PEACE: Covariant learning of nonadiabatic manifolds with parity-resolved Hamiltonians
AI 导读
PEACE 将宇称等变潜空间哈密顿量与可学习的电子连接项结合，用于非绝热分子动力学模拟。消融实验显示，对称性允许的态混合与电子参考系变化在复现交叉结构和弛豫动力学中各有作用；PEACE 能较好复现第一性原理模拟的激发态布居动力学，扩展至自旋轨道耦合后还可模拟系间窜越。
正文
Abstract:Nonadiabatic molecular dynamics provides mechanistic insight into light-driven processes and informs the design of molecules and materials for solar energy conversion, photocatalysis and photo switching. Accurately describing these processes requires a representation that respects electronic symmetry and consistently relates energies to interstate couplings. Here we introduce PEACE, which combines a parity-equivariant latent Hamiltonian with a learned electronic connection. Controlled ablations reveal the complementary roles of symmetry-allowed state mixing and electronic-frame variation in reproducing crossing structures and relaxation dynamics. PEACE closely reproduces excited-state population dynamics from first-principles simulations, while its extension to spin-orbit coupling enables simulations of intersystem crossing. These results demonstrate that a more complete incorporation of the underlying physics into learned electronic representations leads to more accurate predictions of nonadiabatic dynamics.
| Subjects: | Chemical Physics (physics.chem-ph); Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.09576 [physics.chem-ph] |
| (or arXiv:2610.09576v1 [physics.chem-ph] for this version) | |
| https://doi.org/10.48550/arXiv.2610.09576 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Rongzhi Gao [view email]
[v1]
Wed, 7 Oct 2026 07:21:00 UTC (13,909 KB)
来源：arXiv：cs.LG · arxiv.org