Measuring and Suppressing Quantum State Leakage in a Superconducting Qubit

TL;DR

This paper demonstrates methods to measure and significantly reduce leakage errors in superconducting qubits, achieving gate errors below 10^-3 and leakage rates at 10^-5, crucial for quantum error correction.

Contribution

The authors optimize DRAG pulse shaping and detuning to suppress leakage errors in superconducting qubits, providing a practical approach for improving quantum gate fidelity.

Findings

Gate errors below 10^-3 achieved

Leakage rates at 10^-5 level measured

Incoherent heating identified as a major leakage source

Abstract

Leakage errors occur when a quantum system leaves the two-level qubit subspace. Reducing these errors is critically important for quantum error correction to be viable. To quantify leakage errors, we use randomized benchmarking in conjunction with measurement of the leakage population. We characterize single qubit gates in a superconducting qubit, and by refining our use of Derivative Reduction by Adiabatic Gate (DRAG) pulse shaping along with detuning of the pulses, we obtain gate errors consistently below $10^{-3}$ and leakage rates at the $10^{-5}$ level. With the control optimized, we find that a significant portion of the remaining leakage is due to incoherent heating of the qubit.