Inferring low-dimensional microstructure representations using convolutional neural networks
This work addresses the challenge of statistical representation in materials science, offering an incremental improvement over existing methods for analyzing microstructural data.
The researchers tackled the problem of representing microstructural images in materials informatics by using a pre-trained convolutional neural network and manifold learning to create low-dimensional embeddings, showing that this method significantly outperforms traditional two-point correlation methods in extracting image generation parameters.
We apply recent advances in machine learning and computer vision to a central problem in materials informatics: The statistical representation of microstructural images. We use activations in a pre-trained convolutional neural network to provide a high-dimensional characterization of a set of synthetic microstructural images. Next, we use manifold learning to obtain a low-dimensional embedding of this statistical characterization. We show that the low-dimensional embedding extracts the parameters used to generate the images. According to a variety of metrics, the convolutional neural network method yields dramatically better embeddings than the analogous method derived from two-point correlations alone.