Benchmarking Machine Learning Approaches for Polarization Mapping in Ferroelectrics Using 4D-STEM

TL;DR

This paper benchmarks various machine learning models for polarization mapping in ferroelectrics using 4D-STEM data, highlighting challenges in real-world application due to domain gaps and potential for defect detection.

Contribution

It systematically compares multiple ML models for polarization detection in 4D-STEM data and explores strategies to improve transferability from synthetic to experimental data.

Findings

High accuracy on synthetic data models

Domain gap limits real-world applicability

Irregular prediction patterns relate to crystal defects

Abstract

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.