Fully automated identification of 2D material samples

TL;DR

This paper presents an automated neural-network-based system for accurate, efficient identification of 2D material samples, streamlining the complex process of materials classification in quantum research.

Contribution

It introduces a novel iterative neural network approach for classifying imperfect, imbalanced datasets and integrates it into an end-to-end automated data processing pipeline.

Findings

High accuracy in 2D material classification

Effective handling of noisy and imbalanced data sets

Significant reduction in manual identification effort

Abstract

Thin nanomaterials are key constituents of modern quantum technologies and materials research. Identifying specimens of these materials with properties required for the development of state of the art quantum devices is usually a complex and lengthy human task. In this work we provide a neural-network driven solution that allows for accurate and efficient scanning, data-processing and sample identification of experimentally relevant two-dimensional materials. We show how to approach classification of imperfect imbalanced data sets using an iterative application of multiple noisy neural networks. We embed the trained classifier into a comprehensive solution for end-to-end automatized data processing and sample identification.