English

Accelerating Moment Tensor Potentials through Post-Training Pruning

Materials Science 2025-10-23 v1

Abstract

Moment Tensor Potentials (MTPs) are machine-learning interatomic potentials whose basis functions are typically selected using a level-based scheme that is data-agnostic. We introduce a post-training, cost-aware pruning strategy that removes expensive basis functions with minimal loss of accuracy. Applied to nickel and silicon-oxygen systems, it yields models up to seven times faster than standard MTPs. The method requires no new data and remains fully compatible with current MTP implementations.

Keywords

Cite

@article{arxiv.2510.19737,
  title  = {Accelerating Moment Tensor Potentials through Post-Training Pruning},
  author = {Zijian Meng and Karim Zongo and Matthew Thoms and Ryan Eric Grant and Laurent Karim Béland},
  journal= {arXiv preprint arXiv:2510.19737},
  year   = {2025}
}

Comments

16 pages, 5 figures Software available from https://github.com/RichardZJM/MTP_basis_optimization

R2 v1 2026-07-01T07:00:05.256Z