Co-designing Transmon devices for control with simple pulses

TL;DR

This paper uses optimal control to co-design superconducting transmon devices and simple control pulses, aiming to maximize gate fidelity for quantum algorithms in the NISQ era.

Contribution

It introduces a framework combining optimal control and device design to optimize gate performance with simple pulses in superconducting qubits.

Findings

Optimal control can enhance gate fidelity in superconducting qubits.

Design choices significantly impact entangling gate performance.

Simple pulses can achieve high fidelity comparable to complex controls.

Abstract

In the current NISQ era, there is demand for functional quantum devices to solve relevant computational problems, which motivates a utilitarian perspective on device design: The goal is to create a device that is able to run a given algorithm with state-of-the-art performance. In this work, we use optimal control tools to derive the gate set required by a toy algorithm and, in tandem, explore the model space of superconducting quantum computer design, from dispersively coupled to stronger interacting qubits, to maximize gate fidelity. We employ perfect entangler theory to provide flexibility in the search for a two-qubit gate on a given platform and to compare designs with different entangling mechanisms, e.g., $\texttt{CPHASE}$ and $\sqrt{\texttt{iSWAP}}$. To ensure the applicability of our investigation, we limit ourselves to "simple" (i.e., sparse parametrization) pulses and quantify, where results differ from using the full complexity of piecewise constant controls.