Robust quantum dots charge autotuning using neural network uncertainty
arXiv:2406.05175 · doi:10.1088/2632-2153/ad88d5
Abstract
This study presents a machine-learning-based procedure to automate the charge tuning of semiconductor spin qubits with minimal human intervention, addressing one of the significant challenges in scaling up quantum dot technologies. This method exploits artificial neural networks to identify noisy transition lines in stability diagrams, guiding a robust exploration strategy leveraging neural networks' uncertainty estimations. Tested across three distinct offline experimental datasets representing different single quantum dot technologies, the approach achieves over 99% tuning success rate in optimal cases, where more than 10% of the success is directly attributable to uncertainty exploitation. The challenging constraints of small training sets containing high diagram-to-diagram variability allowed us to evaluate the capabilities and limits of the proposed procedure.
12 pages (main) + 14 pages (supplementary)
References in corpus (25)
- On Calibration of Modern Neural Networks
- Weight Uncertainty in Neural Networks
- An addressable quantum dot qubit with fault-tolerant control fidelity
- Semiconductor Spin Qubits
- Computing with spin qubits at the surface code error threshold
- Two-qubit silicon quantum processor with operation fidelity exceeding 99%
- A Survey of Uncertainty in Deep Neural Networks
- Cross-Entropy Loss Functions: Theoretical Analysis and Applications
- Classifier Calibration: A survey on how to assess and improve predicted class probabilities
- Learning Quantum Systems
- A shuttling-based two-qubit logic gate for linking distant silicon quantum processors
- Evaluating model calibration in classification
- Single-electron quantum dot in Si/SiGe with integrated charge-sensing
- Colloquium: Advances in automation of quantum dot devices control
- Low charge noise quantum dots with industrial CMOS manufacturing
- A Machine Learning Approach for Automated Fine-Tuning of Semiconductor Spin Qubits
- Tuning arrays with rays: Physics-informed tuning of quantum dot charge states
- Uniform Spin Qubit Devices in an All-Silicon 300 mm Integrated Process
- An automated approach for consecutive tuning of quantum dot arrays
- Cross-architecture Tuning of Silicon and SiGe-based Quantum Devices Using Machine Learning
- Estimation of Convex Polytopes for Automatic Discovery of Charge State Transitions in Quantum Dot Arrays
- A quantitative study of bias triangles presented in chemical potential space
- A Cryogenic Memristive Neural Decoder for Fault-tolerant Quantum Error Correction
- Dual Operation of Gate-All-Around Silicon Nanowires at Cryogenic Temperatures: FET and Quantum Dot
- Bayesian autotuning of Hubbard model quantum simulators