Combining Variational Autoencoders and Physical Bias for Improved Microscopy Data Analysis
This work addresses the challenge of unsupervised feature analysis in microscopy data for materials science, offering an incremental improvement by integrating physical bias into machine learning methods.
The paper tackles the problem of analyzing microscopy data to identify physically separate regions like phases and boundaries by combining Variational Autoencoders with a physics-driven loss function to minimize discontinuities in latent representations, applied to materials such as NiO-LSMO, BiFeO3, and graphene, demonstrating effectiveness in extracting meaningful information from large imaging datasets.
Electron and scanning probe microscopy produce vast amounts of data in the form of images or hyperspectral data, such as EELS or 4D STEM, that contain information on a wide range of structural, physical, and chemical properties of materials. To extract valuable insights from these data, it is crucial to identify physically separate regions in the data, such as phases, ferroic variants, and boundaries between them. In order to derive an easily interpretable feature analysis, combining with well-defined boundaries in a principled and unsupervised manner, here we present a physics augmented machine learning method which combines the capability of Variational Autoencoders to disentangle factors of variability within the data and the physics driven loss function that seeks to minimize the total length of the discontinuities in images corresponding to latent representations. Our method is applied to various materials, including NiO-LSMO, BiFeO3, and graphene. The results demonstrate the effectiveness of our approach in extracting meaningful information from large volumes of imaging data. The fully notebook containing implementation of the code and analysis workflow is available at https://github.com/arpanbiswas52/PaperNotebooks