Inverse design of bespoke interatomic potentials via active learning by information-matching

2026-06-06

AI Summary

This work addresses the inefficiency of conventional active learning approaches in constructing interatomic potentials, which typically fail to optimize uncertainty with respect to target material properties such as plastic strength. To overcome this limitation, the authors propose an Information Matching (IM) active learning strategy that selects low-cost, intermediate quantities highly correlated with plastic strength as surrogate targets, enabling data-efficient training. The method further incorporates a posterior uncertainty inflation mechanism to correct model errors. By integrating interatomic potential modeling, uncertainty quantification, and molecular simulations, the proposed framework significantly enhances prediction accuracy for both the intermediate quantities and the target plastic strength—even with extremely limited training data—demonstrating the effectiveness and promise of uncertainty-aware active learning for predicting complex material properties.

Abstract

Interatomic potentials (IPs) enable large-scale atomistic simulations beyond the reach of first-principles methods, but their predictive reliability depends critically on the selection of training data, quantified uncertainty, and model expressiveness. Active learning (AL) provides a principled framework for constructing efficient and accurate IPs, yet most strategies reduce parameter uncertainty without explicitly accounting for the specific material properties being predicted. The information-matching (IM) approach addresses this limitation by requiring that the selected training data provide at least as much parameter space information as needed to achieve prescribed uncertainty targets for selected quantities of interest (QoIs). Here, we apply IM to develop bespoke IPs specifically tailored for predicting plastic strength in metals. Due to the high computational cost of simulating plastic strength, we employ an indirect IM strategy that targets inexpensive intermediate QoIs that correlate with strength. The IM method enables precise parameter constraints with minimal training data, yielding precise predictions for both the intermediate QoIs and plastic strength. Yet, model error remains a key limitation, and a post hoc uncertainty inflation correction provides a viable means to mitigate this limitation. These findings illustrate both the promise and limits of uncertainty-aware AL for predicting complex material properties.

Problem

• interatomic potentials
• active learning
• information-matching
• plastic strength
• uncertainty quantification

Innovation

• information-matching
• active learning
• interatomic potentials
• uncertainty quantification
• inverse design