Enabling robust offline active learning for machine learning potentials using simple physics-based priors

TL;DR

This paper introduces a physics-based prior approach for offline active learning in machine learning potentials, significantly reducing the need for extensive data and enabling stable convergence in quantum simulations.

Contribution

It presents a $ riangle$-machine learning method that ensures stable offline active learning by preventing unphysical configurations, improving efficiency and reliability.

Findings

Achieved 70-90% reduction in first-principles calculations.

Demonstrated effectiveness in structural relaxation, transition state, and molecular dynamics.

Enabled stable offline active learning with fewer data points.

Abstract

Machine learning surrogate models for quantum mechanical simulations has enabled the field to efficiently and accurately study material and molecular systems. Developed models typically rely on a substantial amount of data to make reliable predictions of the potential energy landscape or careful active learning and uncertainty estimates. When starting with small datasets, convergence of active learning approaches is a major outstanding challenge which limited most demonstrations to online active learning. In this work we demonstrate a $\Delta$-machine learning approach that enables stable convergence in offline active learning strategies by avoiding unphysical configurations. We demonstrate our framework's capabilities on a structural relaxation, transition state calculation, and molecular dynamics simulation, with the number of first principle calculations being cut down anywhere from 70-90%. The approach is incorporated and developed alongside AMPtorch, an open-source machine learning potential package, along with interactive Google Colab notebook examples.