通过主动学习预测振动谱：4G-NNPs
材料科学
2025-12-04 v2 化学物理
摘要
预测复杂凝聚相和界面系统中振动谱的振动谱，由于拥有数千个自由度，这一任务一直是现代凝聚态理论的挑战。本文开发了利用主动学习和查询-委员会方法构建的第四代高维委员会神经网络势能（4G-HDCNNPs），并将其引入路径积分（PI）分子动力学模拟中的核量子效应（NQEs）、构象熵以及各向异性。通过代表性的液体 bulk 和气-液界面测试案例，我们展示了所开发框架在红外谱模拟中的准确性。 Specifically, by seamlessly integrating non-local charge transfer effects from 4G-HDCNNPs with the NQEs from PI methods, our introduced methodology yields accurate infrared spectra using predicted charges from the 4G-HDCNNP architecture without explicit training of dipole moments. The framework introduced in this work is simple and general, offering a practical paradigm for predictive spectral simulations of complex condensed phases and interfaces, free from empirical parameterizations and ad hoc fitting.
引用
@article{arxiv.2511.01543,
title = {Predictive quantum vibrational spectra through active learning 4G-NNPs},
author = {Md Omar Faruque and Dil K. Limbu and Nathan London and Mohammad R. Momeni},
journal= {arXiv preprint arXiv:2511.01543},
year = {2025}
}