Uncertainty-aware molecular dynamics from Bayesian active learning for phase transformations and thermal transport in SiC

Machine learning interatomic force fields are promising for combining high computational efficiency and accuracy in modeling quantum interactions and simulating atomistic dynamics. Active learning methods have been recently developed to train force fields efficiently and automatically. Among them, Bayesian active learning utilizes principled uncertainty quantification to make data acquisition decisions. In this work, we present a general Bayesian active learning workflow, where the force field is constructed from a sparse Gaussian process regression model based on atomic cluster expansion descriptors. To circumvent the high computational cost of the sparse Gaussian process uncertainty calculation, we formulate a high-performance approximate mapping of the uncertainty and demonstrate a speedup of several orders of magnitude. We demonstrate the autonomous active learning workflow by training a Bayesian force field model for silicon carbide (SiC) polymorphs in only a few days of computer time and show that pressure-induced phase transformations are accurately captured. The resulting model exhibits close agreement with both ab initio calculations and experimental measurements, and outperforms existing empirical models on vibrational and thermal properties. The active learning workflow readily generalizes to a wide range of material systems and accelerates their computational understanding.

Automated summary

Machine learning interatomic force fields combine computational efficiency and accuracy in modeling quantum interactions and simulating atomistic dynamics. Bayesian active learning is a promising approach for efficient training of force fields. This study presents a general Bayesian active learning workflow that utilizes sparse Gaussian process regression and a high-performance approximate mapping to speed up the uncertainty calculation. The resulting model accurately captures pressure-induced phase transformations and outperforms existing models on vibrational and thermal properties. The active learning workflow can be applied to various material systems and accelerates computational understanding.