Neural network based deep learning analysis of semiconductor quantum dot qubits for automated control

TL;DR

This paper presents a CNN-based method to accurately identify and predict disorder in semiconductor quantum dot qubits from charge stability diagrams, enabling automated device tuning and addressing crosstalk issues.

Contribution

It introduces a novel CNN approach that simultaneously predicts multiple disorder parameters in quantum dot systems with high accuracy, improving over previous low-coupling focused methods.

Findings

High prediction accuracy with R^2 > 0.994 for disorder parameters

Ability to tune five or more quantum dots simultaneously

Provides a validation method for neural network predictions

Abstract

Machine learning offers a largely unexplored avenue for improving noisy disordered devices in physics using automated algorithms. Through simulations that include disorder in physical devices, particularly quantum devices, there is potential to learn about disordered landscapes and subsequently tune devices based on those insights. In this work, we introduce a novel methodology that employs machine learning, specifically convolutional neural networks (CNNs), to discern the disorder landscape in the parameters of the disordered extended Hubbard model underlying the semiconductor quantum dot spin qubit architectures. This technique takes advantage of experimentally obtainable charge stability diagrams from neighboring quantum dot pairs, enabling the CNN to accurately identify disorder in each parameter of the extended Hubbard model. Remarkably, our CNN can process site-specific disorder in Hubbard parameters, including variations in hopping constants, on-site potentials (gate voltages), and both intra-site and inter-site Coulomb terms. This advancement facilitates the prediction of spatially dependent disorder across all parameters simultaneously with high accuracy ($R^2>0.994$) and fewer parameter constraints, marking a significant improvement over previous methods that were focused only on analyzing on-site potentials at low coupling. Furthermore, our approach allows for the tuning of five or more quantum dots at a time, effectively addressing the often-overlooked issue of crosstalk. Not only does our method streamline the tuning process, potentially enabling fully automated adjustments, but it also introduces a "no trust" verification method to rigorously validate the neural network's predictions. Ultimately, this work aims to lay the groundwork for generalizing our method to tackle a broad spectrum of physical problems.