Saved in:
Bibliographic Details
Main Authors: Chen, Ruiyang, Zhang, Qingyuan, Chen, Ji
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2604.03027
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911565084622848
author Chen, Ruiyang
Zhang, Qingyuan
Chen, Ji
author_facet Chen, Ruiyang
Zhang, Qingyuan
Chen, Ji
contents Machine learning force field (MLFF) has emerged as a powerful data-driven tool for atomistic simulations, enabling large-scale and complex atomic systems to be simulated with accuracy comparable to \textit{ab initio} methods. However, MLFFs often suffer from low training efficiency in the phase transition regime, where structural fluctuations are significantly elevated. To address this challenge, we propose a Central-Peripheral Distillation (CPD) algorithm for training dataset distillation. By strategically integrating representative samples with critical corner cases, the CPD algorithm ensures that the distilled dataset retains maximum structural diversity. We validated the efficacy of the CPD method on the liquid-liquid phase transition of dense hydrogen. Results show that, with the CPD approach, only 200 configurations are sufficient to train a MLFF that can fully reproduce the structural and dynamical properties of liquid hydrogen in the vicinity of its phase transition regime. This work paves the way for high-fidelity labeling of the MLFF training datasets, for instance by adopting high-level \textit{ab initio} calculations beyond the standard density functional theory, thereby enhancing the predictive accuracy of MLFFs.
format Preprint
id arxiv_https___arxiv_org_abs_2604_03027
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Dataset Distillation for Machine Learning Force Field in Phase Transition Regime
Chen, Ruiyang
Zhang, Qingyuan
Chen, Ji
Chemical Physics
Machine learning force field (MLFF) has emerged as a powerful data-driven tool for atomistic simulations, enabling large-scale and complex atomic systems to be simulated with accuracy comparable to \textit{ab initio} methods. However, MLFFs often suffer from low training efficiency in the phase transition regime, where structural fluctuations are significantly elevated. To address this challenge, we propose a Central-Peripheral Distillation (CPD) algorithm for training dataset distillation. By strategically integrating representative samples with critical corner cases, the CPD algorithm ensures that the distilled dataset retains maximum structural diversity. We validated the efficacy of the CPD method on the liquid-liquid phase transition of dense hydrogen. Results show that, with the CPD approach, only 200 configurations are sufficient to train a MLFF that can fully reproduce the structural and dynamical properties of liquid hydrogen in the vicinity of its phase transition regime. This work paves the way for high-fidelity labeling of the MLFF training datasets, for instance by adopting high-level \textit{ab initio} calculations beyond the standard density functional theory, thereby enhancing the predictive accuracy of MLFFs.
title Dataset Distillation for Machine Learning Force Field in Phase Transition Regime
topic Chemical Physics
url https://arxiv.org/abs/2604.03027