Evaluating the diversity and utility of materials proposed by generative models
This work identifies critical issues in using generative models for materials science, which could hinder practical applications like discovering new materials.
The study evaluated the physics-guided crystal generation model (PGCGM) for inverse design of materials, finding that its input space lacks smoothness for optimization and most generated structures are thermodynamically unstable, highlighting limitations in current generative approaches.
Generative machine learning models can use data generated by scientific modeling to create large quantities of novel material structures. Here, we assess how one state-of-the-art generative model, the physics-guided crystal generation model (PGCGM), can be used as part of the inverse design process. We show that the default PGCGM's input space is not smooth with respect to parameter variation, making material optimization difficult and limited. We also demonstrate that most generated structures are predicted to be thermodynamically unstable by a separate property-prediction model, partially due to out-of-domain data challenges. Our findings suggest how generative models might be improved to enable better inverse design.