Reduced phase error through optimized control of a superconducting qubit

TL;DR

This paper introduces amplified phase error pulses and optimized control techniques to significantly reduce phase errors and leakage in superconducting qubits, enhancing quantum gate fidelity.

Contribution

The authors develop a new metrology for phase error detection and implement a simplified DRAG control method to minimize errors in superconducting qubits.

Findings

Phase errors reduced by about five times to ~1.6° per gate.

Leakage outside the qubit manifold decreased to ~10^{-4}.

Gate speed increased by 20% without increasing errors.

Abstract

Minimizing phase and other errors in experimental quantum gates allows higher fidelity quantum processing. To quantify and correct for phase errors in particular, we have developed a new experimental metrology --- amplified phase error (APE) pulses --- that amplifies and helps identify phase errors in general multi-level qubit architectures. In order to correct for both phase and amplitude errors specific to virtual transitions and leakage outside of the qubit manifold, we implement "half derivative" an experimental simplification of derivative reduction by adiabatic gate (DRAG) control theory. The phase errors are lowered by about a factor of five using this method to $\sim 1.6^{\circ}$ per gate, and can be tuned to zero. Leakage outside the qubit manifold, to the qubit $|2\rangle$ state, is also reduced to $\sim 10^{-4}$ for $20\%$ faster gates.