Curiosity Driven Exploration to Optimize Structure-Property Learning in Microscopy
This provides a more efficient tool for materials scientists to accelerate structure-property learning in microscopy, though it is incremental compared to existing deep kernel active learning methods.
The paper tackled the problem of efficiently mapping structure-property correlations in microscopy by developing a lightweight curiosity algorithm that actively samples unexplored regions, outperforming random sampling for property prediction.
Rapidly determining structure-property correlations in materials is an important challenge in better understanding fundamental mechanisms and greatly assists in materials design. In microscopy, imaging data provides a direct measurement of the local structure, while spectroscopic measurements provide relevant functional property information. Deep kernel active learning approaches have been utilized to rapidly map local structure to functional properties in microscopy experiments, but are computationally expensive for multi-dimensional and correlated output spaces. Here, we present an alternative lightweight curiosity algorithm which actively samples regions with unexplored structure-property relations, utilizing a deep-learning based surrogate model for error prediction. We show that the algorithm outperforms random sampling for predicting properties from structures, and provides a convenient tool for efficient mapping of structure-property relationships in materials science.