Data-Efficient Machine Learning Potentials via Difference Vectors Based on Local Atomic Environments

TL;DR

This paper introduces DV-LAE, a method that uses difference vectors based on local atomic environments to optimize datasets for machine learning potentials, reducing redundancy and training time while maintaining accuracy.

Contribution

The paper presents a novel histogram-based descriptor method, DV-LAE, for dataset optimization and out-of-distribution detection in atomistic machine learning potentials.

Findings

Reduces dataset size by up to 56% while maintaining accuracy.

Cuts training time per iteration by over 50%.

Enables visualization for out-of-distribution detection.

Abstract

Constructing efficient and diverse datasets is essential for the development of accurate machine learning potentials (MLPs) in atomistic simulations. However, existing approaches often suffer from data redundancy and high computational costs. Herein, we propose a new method--Difference Vectors based on Local Atomic Environments (DV-LAE)--that encodes structural differences via histogram-based descriptors and enables visual analysis through t-SNE dimensionality reduction. This approach facilitates redundancy detection and dataset optimization while preserving structural diversity. We demonstrate that DV-LAE significantly reduces dataset size and training time across various materials systems, including high-pressure hydrogen, iron-hydrogen binaries, magnesium hydrides, and carbon allotropes, with minimal compromise in prediction accuracy. For instance, in the $\alpha$-Fe/H system, maintaining a highly similar MLP accuracy, the dataset size was reduced by 56%, and the training time per iteration dropped by over 50%. Moreover, we show how visualizing the DV-LAE representation aids in identifying out-of-distribution data by examining the spatial distribution of high-error prediction points, providing a robust reliability metric for new structures during simulations. Our results highlight the utility of local environment visualization not only as an interpretability tool but also as a practical means for accelerating MLP development and ensuring data efficiency in large-scale atomistic modeling.