Learning Large-Time-Step Molecular Dynamics with Graph Neural Networks

Abstract

Molecular dynamics (MD) simulation predicts the trajectory of atoms by solving Newton's equation of motion with a numeric integrator. Due to physical constraints, the time step of the integrator need to be small to maintain sufficient precision. This limits the efficiency of simulation. To this end, we introduce a graph neural network (GNN) based model, MDNet, to predict the evolution of coordinates and momentum with large time steps. In addition, MDNet can easily scale to a larger system, due to its linear complexity with respect to the system size. We demonstrate the performance of MDNet on a 4000-atom system with large time steps, and show that MDNet can predict good equilibrium and transport properties, well aligned with standard MD simulations.

PDF NeurIPSW OpenReview Semantic Scholar

Cite

Text

Zheng et al. "Learning Large-Time-Step Molecular Dynamics with Graph Neural Networks." NeurIPS 2021 Workshops: AI4Science, 2021.

Markdown

[Zheng et al. "Learning Large-Time-Step Molecular Dynamics with Graph Neural Networks." NeurIPS 2021 Workshops: AI4Science, 2021.](https://mlanthology.org/neuripsw/2021/zheng2021neuripsw-learning/)

BibTeX

@inproceedings{zheng2021neuripsw-learning,
  title     = {{Learning Large-Time-Step Molecular Dynamics with Graph Neural Networks}},
  author    = {Zheng, Tianze and Gao, Weihao and Wang, Chong},
  booktitle = {NeurIPS 2021 Workshops: AI4Science},
  year      = {2021},
  url       = {https://mlanthology.org/neuripsw/2021/zheng2021neuripsw-learning/}
}