Reward driven workflows for unsupervised explainable analysis of phases and ferroic variants from atomically resolved imaging data

TL;DR

This paper introduces a reward-driven workflow for optimizing unsupervised machine learning methods to analyze atomic-scale imaging data, enabling better identification of material phases and variants by aligning analysis with physical behaviors.

Contribution

It presents a novel reward-driven hyperparameter optimization approach for unsupervised analysis of materials imaging data, improving the discovery of physical features and disentangling structural variations.

Findings

Optimized hyperparameters improve phase and variant identification.

Reward-driven approach aligns analysis with physical domain knowledge.

Extended workflow with variational autoencoder enhances structural disentanglement.

Abstract

Rapid progress in aberration corrected electron microscopy necessitates development of robust methods for the identification of phases, ferroic variants, and other pertinent aspects of materials structure from imaging data. While unsupervised methods for clustering and classification are widely used for these tasks, their performance can be sensitive to hyperparameter selection in the analysis workflow. In this study, we explore the effects of descriptors and hyperparameters on the capability of unsupervised ML methods to distill local structural information, exemplified by discovery of polarization and lattice distortion in Sm doped BiFeO3 (BFO) thin films. We demonstrate that a reward-driven approach can be used to optimize these key hyperparameters across the full workflow, where rewards were designed to reflect domain wall continuity and straightness, ensuring that the analysis aligns with the material's physical behavior. This approach allows us to discover local descriptors that are best aligned with the specific physical behavior, providing insight into the fundamental physics of materials. We further extend the reward driven workflows to disentangle structural factors of variation via optimized variational autoencoder (VAE). Finally, the importance of well-defined rewards was explored as a quantifiable measure of success of the workflow.