Weakly Supervised Graph Neural Networks for Scalable 3D Phase Segmentation in Molecular Dynamics Simulations

Abin Shakya, Bijaya Karki

2025

Abstract

Accurate phase identification in large-scale molecular dynamics simulation remains a significant challenge due to ambiguous boundaries between compositionally distinct regions and the lack of ground truth labels. While unsupervised methods can perform phase segmentation for small systems through structure-aware segmentation pipelines, their computational cost becomes prohibitive for large-scale analysis. We present a weakly-supervised machine learning pipeline that trains Graph Neural Networks (GNNs) to enable scalable phase segmentation in 3D atomistic systems. Using a physically grounded unsupervised method, we generate weak labels for small FeMgSiON systems that exhibit Fe-rich (metallic) and Fe-poor (silicate) phase separation. These labels guide GNNs to learn physically meaningful representations of atomic neighborhoods. Once trained, the GNNs act as an efficient parametric model, enabling direct segmentation of arbitrarily large atomistic systems eliminating the computational overhead of the initial unsupervised pipeline. By learning from thousands of weakly labeled snapshots, the model discerns latent structural patterns, enhancing both prediction accuracy and generalization to unseen data. This methodology enables efficient, accurate, and physically consistent phase segmentation in large-scale molecular dynamics, unlocking new possibilities for scalable analysis in material simulations.

Paper Citation

in Harvard Style

Shakya A. and Karki B. (2025). Weakly Supervised Graph Neural Networks for Scalable 3D Phase Segmentation in Molecular Dynamics Simulations. In Proceedings of the 17th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management - Volume 1: KDIR; ISBN , SciTePress, pages 302-312. DOI: 10.5220/0013709400004000

in Bibtex Style

@conference{kdir25,
author={Abin Shakya and Bijaya Karki},
title={Weakly Supervised Graph Neural Networks for Scalable 3D Phase Segmentation in Molecular Dynamics Simulations},
booktitle={Proceedings of the 17th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management - Volume 1: KDIR},
year={2025},
pages={302-312},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013709400004000},
isbn={},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 17th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management - Volume 1: KDIR
TI - Weakly Supervised Graph Neural Networks for Scalable 3D Phase Segmentation in Molecular Dynamics Simulations
SN -
AU - Shakya A.
AU - Karki B.
PY - 2025
SP - 302
EP - 312
DO - 10.5220/0013709400004000
PB - SciTePress