English

Bonding-aware Materials Representation for Deep Learning Atomistic Models

Materials Science 2023-06-19 v2 Computational Physics

Abstract

Deep potentials for molecular dynamics (MD) achieve first-principles accuracy at much lower computational cost. However, their use in large length- and time-scale simulations is limited by their lower speeds compared to analytical atomistic potentials, primarily due to network complexity and long embedding time. Here, based on the moments theorem, we develop a chemical-bonding-aware embedding for neural network potentials that achieve state-of-the-art accuracy in forces and local electronic density of states prediction with an ultrasmall 16x32 neural network resulting in significantly lower computational cost.

Keywords

Cite

@article{arxiv.2306.08285,
  title  = {Bonding-aware Materials Representation for Deep Learning Atomistic Models},
  author = {Or Shafir and Ilya Grinberg},
  journal= {arXiv preprint arXiv:2306.08285},
  year   = {2023}
}
R2 v1 2026-06-28T11:04:41.729Z