Quantum optimal control of superconducting qubits based on machine-learning characterization

TL;DR

This paper introduces a machine-learning-based method to accurately model superconducting qubit dynamics for optimized quantum control, achieving high-fidelity single-qubit gates through tailored pulse design.

Contribution

It presents a novel approach combining physics-inspired machine learning with experimental data to improve quantum control accuracy in superconducting qubits.

Findings

Achieved approximately 99.99% fidelity in single-qubit gates.

Demonstrated the method's effectiveness through detailed numerical simulations.

Provided a practical framework for device-specific quantum control optimization.

Abstract

Implementing fast and high-fidelity quantum operations using open-loop quantum optimal control relies on having an accurate model of the quantum dynamics. Any deviations between this model and the complete dynamics of the device, such as the presence of spurious modes or pulse distortions, can degrade the performance of optimal controls in practice. Here, we propose an experimentally simple approach to realize optimal quantum controls tailored to the device parameters and environment while specifically characterizing this quantum system. Concretely, we use physics-inspired machine learning to infer an accurate model of the dynamics from experimentally available data and then optimize our experimental controls on this trained model. We show the power and feasibility of this approach by optimizing arbitrary single-qubit operations on a superconducting transmon qubit, using detailed numerical simulations. We demonstrate that this framework produces an accurate description of the device dynamics under arbitrary controls, together with the precise pulses achieving arbitrary single-qubit gates with a high fidelity of about 99.99%.