End-to-End Analysis of Charge Stability Diagrams with Transformers

TL;DR

This paper introduces a transformer-based end-to-end model for analyzing charge stability diagrams in quantum dot arrays, improving accuracy, speed, and generalizability over traditional CNN approaches, aiding quantum device control.

Contribution

It applies object detection transformers to quantum dot charge diagrams, demonstrating superior performance and reduced complexity compared to existing methods, without retraining across architectures.

Findings

Outperforms CNNs on three quantum dot architectures

Reduces analysis complexity and runtime

Enhances generalizability across devices

Abstract

Transformer models and end-to-end learning frameworks are rapidly revolutionizing the field of artificial intelligence. In this work, we apply object detection transformers to analyze charge stability diagrams in semiconductor quantum dot arrays, a key task for achieving scalability with spin-based quantum computing. Specifically, our model identifies triple points and their connectivity, which is crucial for virtual gate calibration, charge state initialization, drift correction, and pulse sequencing. We show that it surpasses convolutional neural networks in performance on three different spin qubit architectures, all without the need for retraining. In contrast to existing approaches, our method significantly reduces complexity and runtime, while enhancing generalizability. The results highlight the potential of transformer-based end-to-end learning frameworks as a foundation for a scalable, device- and architecture-agnostic tool for control and tuning of quantum dot devices.