Litcius/Paper detail

Active learning meets metadynamics: automated workflow for reactive machine learning interatomic potentials

Valdas Vitartas, Hanwen Zhang, Veronika Jurásková, Tristan Johnston-Wood, Fernanda Duarte

2025Digital Discovery9 citationsDOIOpen Access PDF

Abstract

2 reaction between fluoride and chloromethane in implicit water, the methyl shift of 2,2-dimethylisoindene in the gas phase, and a glycosylation reaction in explicit dichloromethane solution, where competitive pathways exist. The proposed training strategy yields accurate and stable MLIPs for all three cases, highlighting its versatility for modelling reactive processes.

Topics & Concepts

MetadynamicsComputer scienceWorkflowActive learning (machine learning)Molecular dynamicsMachine learningArtificial intelligenceCheminformaticsQM/MMInteratomic potentialPotential energyTraining setChloromethaneChemistrySample (material)Reaction dynamicsEnergy (signal processing)InitializationBiological systemEnsemble learningArtificial neural networkMachine Learning in Materials ScienceAdvanced Chemical Physics StudiesInorganic Chemistry and Materials
Active learning meets metadynamics: automated workflow for reactive machine learning interatomic potentials | Litcius