English

Accelerating Electron Dynamics Simulations through Machine Learned Time Propagators

Materials Science 2024-07-29 v2 Machine Learning Computational Physics

Abstract

Time-dependent density functional theory (TDDFT) is a widely used method to investigate electron dynamics under various external perturbations such as laser fields. In this work, we present a novel approach to accelerate real time TDDFT based electron dynamics simulations using autoregressive neural operators as time-propagators for the electron density. By leveraging physics-informed constraints and high-resolution training data, our model achieves superior accuracy and computational speed compared to traditional numerical solvers. We demonstrate the effectiveness of our model on a class of one-dimensional diatomic molecules. This method has potential in enabling real-time, on-the-fly modeling of laser-irradiated molecules and materials with varying experimental parameters.

Keywords

Cite

@article{arxiv.2407.09628,
  title  = {Accelerating Electron Dynamics Simulations through Machine Learned Time Propagators},
  author = {Karan Shah and Attila Cangi},
  journal= {arXiv preprint arXiv:2407.09628},
  year   = {2024}
}

Comments

9 pages, 5 figures

R2 v1 2026-06-28T17:39:17.374Z