Fully automated identification of 2D material samples

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.

Assembly of van der Waals heterostructures: exfoliation, searching, and stacking of 2D materials

“…Further forward, deep learning enables the segmentation of various 2D materials. [54][55][56][57] Through training with many flake images, the neural network develops a skill to identify atomic layers on SiO 2 /Si substrates. The algorithm based on deep learning is implemented in the automated searching system 41) described in Sect.…”

Automated tuning of double quantum dots into specific charge states using neural networks

“…Applying machine learning techniques for parameter estimation and tuning of quantum systems has been a promising avenue within this endeavor [22][23][24][25][26][27]. Machine learning methods can be used to automate tasks previously done by humans [28] and construct high-quality abstract models interpreting complex measurements. They are * geliska@phys.ethz.ch fast to evaluate even without implementing a device-or system-specific physical model, which can be complex and lengthy to simulate in the case of QD qubits.…”