English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Machine learning interatomic force fields are promising for combining highcomputational efficiency and accuracy in modeling quantum interactions andsimulating atomistic dynamics. Active learning methods have been recentlydeveloped to train force fields efficiently and automatically. Among them,Bayesian active learning utilizes principled uncertainty quantification to makedata acquisition decisions. In this work, we present a general Bayesian activelearning workflow, where the force field is constructed from a sparse Gaussianprocess regression model based on atomic cluster expansion descriptors. Tocircumvent the high computational cost of the sparse Gaussian processuncertainty calculation, we formulate a high-performance approximate mapping ofthe uncertainty and demonstrate a speedup of several orders of magnitude. Wedemonstrate the autonomous active learning workflow by training a Bayesianforce field model for silicon carbide (SiC) polymorphs in only a few days ofcomputer time and show that pressure-induced phase transformations areaccurately captured. The resulting model exhibits close agreement with both\textit{ab initio} calculations and experimental measurements, and outperformsexisting empirical models on vibrational and thermal properties. The activelearning workflow readily generalizes to a wide range of material systems andaccelerates their computational understanding.

Uncertainty-aware molecular dynamics from Bayesian active learning for Phase Transformations and Thermal Transport in SiC