Quantifying the performance of machine learning models in materials discovery
Christopher K. H. Borg, Eric S. Muckley, Clara Nyby, James E. Saal, Logan Ward, Apurva Mehta, Bryce Meredig
Abstract
In this work, we simulate a sequential learning (SL)-guided materials discovery process and demonstrate a decoupling between traditional model error metrics and model performance in guiding materials discoveries.
Topics & Concepts
Machine learningComputer scienceArtificial intelligenceData scienceMachine Learning in Materials ScienceX-ray Diffraction in CrystallographyComputational Drug Discovery Methods