Decoratypes: An Extensible Crystal Taxonomy for Machine Learning-Guided Materials Discovery

Abstract

We introduce decoratypes as a structure taxonomy that classifies compounds based on site decorations of specific structural prototypes. Building on this foundation, a ferroelectric materials discovery framework is developed, integrating decoratypes with an active learning approach to accelerate exploration. In addition, six novel ferroelectric candidates are predicted, including three strain-activated ferroelectrics and three strain-activated hyperferroelectrics. These findings highlight the potential of the decoratype taxonomy to enhance our understanding of structure-driven material properties and facilitate the discovery of promising yet underexplored regions of chemical space.

15 pages, 16 figures

References in corpus (7)

• Insightful classification of crystal structures using deep learning
• Magnetic octupoles as the order parameter for unconventional antiferromagnetism
• Multipole classification in 122 magnetic point groups for unified understanding of multiferroic responses and transport phenomena
• Two-dimensional ferroelectrics from high throughput computational screening
• Generative Discovery of Novel Chemical Designs using Diffusion Modeling and Transformer Deep Neural Networks with Application to Deep Eutectic Solvents
• Hidden orders and (anti-)Magnetoelectric Effects in Cr${}_2$O${}_3$ and $\alpha$-Fe${}_2$O${}_3$
• Pressure induced metallization of Cu3N