Physics in the Machine: Integrating Physical Knowledge in Autonomous Phase-Mapping

TL;DR

This paper explores how integrating physical knowledge, such as ab-initio phase boundary data, into AI-driven autonomous systems enhances materials phase-mapping and discovery processes.

Contribution

It demonstrates the benefits of incorporating prior physical knowledge into autonomous AI algorithms for phase-mapping in materials science.

Findings

Incorporating physical priors improves phase-mapping efficiency.

Using ab-initio data accelerates discovery of optimal materials.

Physical knowledge integration enhances autonomous exploration.

Abstract

Application of artificial intelligence (AI), and more specifically machine learning, to the physical sciences has expanded significantly over the past decades. In particular, science-informed AI, also known as scientific AI or inductive bias AI, has grown from a focus on data analysis to now controlling experiment design, simulation, execution and analysis in closed-loop autonomous systems. The CAMEO (closed-loop autonomous materials exploration and optimization) algorithm employs scientific AI to address two tasks: learning a material system's composition-structure relationship and identifying materials compositions with optimal functional properties. By integrating these, accelerated materials screening across compositional phase diagrams was demonstrated, resulting in the discovery of a best-in-class phase change memory material. Key to this success is the ability to guide subsequent measurements to maximize knowledge of the composition-structure relationship, or phase map. In this work we investigate the benefits of incorporating varying levels of prior physical knowledge into CAMEO's autonomous phase-mapping. This includes the use of ab-initio phase boundary data from the AFLOW repositories, which has been shown to optimize CAMEO's search when used as a prior.