Data Curation for Machine Learning Interatomic Potentials by Determinantal Point Processes

2026-03-23

AI Summary

This work addresses the high cost of data generation and labeling in training machine learning interatomic potential models by proposing an efficient strategy for selecting informative and diverse atomic configurations. For the first time, determinantal point processes (DPPs) are introduced into potential model data curation, leveraging molecular descriptors to construct a kernel function that enables unsupervised selection of representative subsets from large configuration spaces for quantum mechanical labeling. The approach naturally accommodates heterogeneous and multimodal data and integrates seamlessly into online active learning workflows within molecular dynamics simulations. Experiments on hafnium oxide demonstrate that training sets selected via DPP significantly enhance model accuracy and robustness compared to existing selection strategies.

Technology Category

• Machine Learning: Active Learning
• Reasoning under Uncertainty: Relational Probabilistic Models
• Data Mining & Knowledge Management: Intelligent Query Processing

Application Category

• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics
• Web Mining and Content Analysis: Machine learning and data science for the Web
• User Modeling, Personalization and Recommendation: ML for personalized search and recommendations

Abstract

The development of machine learning interatomic potentials faces a critical computational bottleneck with the generation and labeling of useful training datasets. We present a novel application of determinantal point processes (DPPs) to the task of selecting informative subsets of atomic configurations to label with reference energies and forces from costly quantum mechanical methods. Through experiments with hafnium oxide data, we show that DPPs are competitive with existing approaches to constructing compact but diverse training sets by utilizing kernels of molecular descriptors, leading to improved accuracy and robustness in machine learning representations of molecular systems. Our work identifies promising directions to employ DPPs for unsupervised training data curation with heterogeneous or multimodal data, or in online active learning schemes for iterative data augmentation during molecular dynamics simulation.

Problem

• data curation
• machine learning interatomic potentials
• training data selection
• atomic configurations
• quantum mechanical labeling

Innovation

• Determinantal Point Processes
• Data Curation
• Machine Learning Interatomic Potentials
• Active Learning
• Molecular Descriptors