Developing optimal strategies to calibrate quantum processors for high-fidelity operation is one of the outstanding challenges in quantum computing today. Here, we demonstrate multiple examples of high-fidelity operations achieved using a unified global optimization-driven automated calibration routine on a six dot semiconductor quantum processor. Within the same algorithmic framework we optimize readout, shuttling and single-qubit quantum gates by tailoring task-specific cost functions and tuning parameters based on the underlying physics of each operation. Our approach reaches systematically 99% readout fidelity, >99% shuttling fidelity over an effective distance of 10μm, and >99.5% single-qubit gate fidelity on timescales similar or shorter compared to those of expert human operators. The flexibility of our gradient-free closed loop algorithmic procedure allows for seamless application across diverse qubit functionalities while providing a systematic framework to tune-up semiconductor quantum devices and enabling interpretability of the identified optimal operation points.
@article{arxiv.2503.12256,
title = {Unified evolutionary optimization for high-fidelity spin qubit operations},
author = {Sam R. Katiraee-Far and Yuta Matsumoto and Brennan Undseth and Maxim De Smet and Valentina Gualtieri and Christian Ventura Meinersen and Irene Fernandez de Fuentes and Kenji Capannelli and Maximilian Rimbach-Russ and Giordano Scappucci and Lieven M. K. Vandersypen and Eliska Greplova},
journal= {arXiv preprint arXiv:2503.12256},
year = {2025}
}