Tag
This paper introduces a budget-dependent method that combines coverage-based and response-based training-set selection strategies for machine-learned interatomic potentials, aiming to improve accuracy and efficiency in materials simulations.