PyAPX: Python toolkit for atomic configuration pattern exploration

TL;DR

PyAPX is a Python toolkit that uses Bayesian methods and novel encoding techniques to efficiently explore atomic configurations in crystalline materials, aiding detailed material design.

Contribution

The paper introduces PyAPX, a toolkit with new encoding methods for configuration search, demonstrating improved convergence in materials discovery tasks.

Findings

Encoding methods in PyAPX outperform traditional one-hot encoding.

PyAPX effectively explores atomic configurations in crystalline materials.

The toolkit is broadly applicable to materials discovery processes.

Abstract

In materials discovery, the integration of first-principles calculations with machine learning techniques has been actively studied for two key tasks: crystal structure prediction, which searches for stable structures given a chemical composition, and elemental substitution, which explores chemical compositions that yield desirable properties in a given crystal structure. However, even when both the crystal structure and chemical composition are fixed, material properties can still vary depending on the atomic arrangements (configurations) at crystallographic sites. To support detailed material design, we present PyAPX, a Python toolkit that performs Bayesian searches of stable atomic configurations. A distinctive feature of this initial release is the introduction of encoding methods suitable for configuration search, and we evaluate their performance using the h-BCN system. As a result, they were confirmed to yield superior convergence compared to commonly used one-hot encoding. PyAPX is broadly applicable to crystalline materials and is expected to further advance materials discovery.