Building Data-driven Models with Microstructural Images: Generalization and Interpretability

TL;DR

This paper investigates convolutional neural networks for microstructural image classification, emphasizing generalization, feature efficiency, and interpretability to enhance understanding of process-structure-property relationships in materials science.

Contribution

It introduces a comprehensive approach to using CNNs for microstructure analysis, focusing on model generalization, feature reduction, and interpretability beyond accuracy.

Findings

CNN models can generalize across different microstructural datasets.

Effective feature reduction is achievable without sacrificing performance.

Interpretability methods reveal meaningful microstructural features.

Abstract

As data-driven methods rise in popularity in materials science applications, a key question is how these machine learning models can be used to understand microstructure. Given the importance of process-structure-property relations throughout materials science, it seems logical that models that can leverage microstructural data would be more capable of predicting property information. While there have been some recent attempts to use convolutional neural networks to understand microstructural images, these early studies have focused only on which featurizations yield the highest machine learning model accuracy for a single data set. This paper explores the use of convolutional neural networks for classifying microstructure with a more holistic set of objectives in mind: generalization between data sets, number of features required, and interpretability.