Unsupervised Atomic Data Mining via Multi-Kernel Graph Autoencoders for Machine Learning Force Fields
Authors:
Preprint
Abstract
Constructing a chemically diverse dataset while avoiding sampling bias is critical to training efficient and generalizable force fields. However, in computational chemistry and materials science, many common dataset generation techniques are prone to oversampling regions of the potential energy surface. Furthermore, these regions can be difficult to identify and isolate from each other or may not align well with human intuition, making it challenging to systematically remove bias in the dataset. While traditional clustering and pruning (down-sampling) approaches can be useful for this, they can often lead to information loss or a failure to properly identify distinct regions of the potential energy surface due to difficulties associated with the high dimensionality of atomic descriptors. In this work, we introduce the Multi-kernel Edge Attention-based Graph Autoencoder (MEAGraph) model, an unsupervised approach for analyzing atomic datasets. MEAGraph combines multiple linear kernel transformations with attention-based message passing to capture geometric sensitivity and enable effective dataset pruning without relying on labels or extensive training. Demonstrated applications on niobium, tantalum, and iron datasets show that MEAGraph efficiently groups similar atomic environments, allowing for the use of basic pruning techniques for removing sampling bias. This approach provides an effective method for representation learning and clustering that can be used for data analysis, outlier detection, and dataset optimization.Keywords
All Related Versions
This publication has 19 references indexed in Scilit:
- Model-free estimation of completeness, uncertainties, and outliers in atomistic machine learning using information theoryNature Communications, 2025
- Exploring structural transitions at grain boundaries in Nb using a generalized embedded atom interatomic potentialComputational Materials Science, 2023
- Uncertainty-driven dynamics for active learning of interatomic potentialsNature Computational Science, 2023
- Training data selection for accuracy and transferability of interatomic potentialsnpj Computational Materials, 2022
- E(3)-equivariant graph neural networks for data-efficient and accurate interatomic potentialsNature Communications, 2022
- Data-driven simulation and characterisation of gold nanoparticle meltingNature Communications, 2021
- Differentiable sampling of molecular geometries with uncertainty-based adversarial attacksNature Communications, 2021
- Metadynamics sampling in atomic environment space for collecting training data for machine learning potentialsnpj Computational Materials, 2021
- wACSF—Weighted atom-centered symmetry functions as descriptors in machine learning potentialsThe Journal of Chemical Physics, 2018
- Quantum-chemical insights from deep tensor neural networksNature Communications, 2017