Data Curation for Machine Learning Interatomic Potentials by Determinantal Point Processes

TL;DR

This paper introduces a novel use of determinantal point processes for selecting diverse and informative atomic configurations to efficiently generate training datasets for machine learning interatomic potentials, improving accuracy and robustness.

Contribution

It applies DPPs to the task of data curation in molecular simulations, demonstrating improved dataset diversity and model performance over existing methods.

Findings

DPPs are competitive with existing data selection methods.

Using molecular descriptor kernels enhances diversity and accuracy.

DPPs enable promising directions for active learning and data augmentation.

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.