Benchmarking Machine Learning Approaches for Polarization Mapping in Ferroelectrics Using 4D-STEM
This work addresses the challenge of extracting polarization directions from 4D-STEM data for ferroelectrics, which is incremental as it benchmarks existing methods and highlights barriers to real-world application.
The study benchmarked machine learning models like ResNet and VGG to automate polarization direction detection from 4D-STEM diffraction patterns in ferroelectrics, finding that synthetic data training achieved high accuracy but faced a domain gap with experiments, and error patterns indicated limitations in classification and potential for defect detection.
Four-dimensional scanning transmission electron microscopy (4D-STEM) provides rich, atomic-scale insights into materials structures. However, extracting specific physical properties - such as polarization directions essential for understanding functional properties of ferroelectrics - remains a significant challenge. In this study, we systematically benchmark multiple machine learning models, namely ResNet, VGG, a custom convolutional neural network, and PCA-informed k-Nearest Neighbors, to automate the detection of polarization directions from 4D-STEM diffraction patterns in ferroelectric potassium sodium niobate. While models trained on synthetic data achieve high accuracy on idealized synthetic diffraction patterns of equivalent thickness, the domain gap between simulation and experiment remains a critical barrier to real-world deployment. In this context, a custom made prototype representation training regime and PCA-based methods, combined with data augmentation and filtering, can better bridge this gap. Error analysis reveals periodic missclassification patterns, indicating that not all diffraction patterns carry enough information for a successful classification. Additionally, our qualitative analysis demonstrates that irregularities in the model's prediction patterns correlate with defects in the crystal structure, suggesting that supervised models could be used for detecting structural defects. These findings guide the development of robust, transferable machine learning tools for electron microscopy analysis.