Abstract
We introduce a Bayesian method for the estimation of single qubit errors in quantum devices, and use it to characterize these errors on three 27-qubit superconducting qubit devices. We self-consistently estimate up to seven parameters of each qubit's state preparation, readout, and gate errors, analyze the stability of these errors as a function of time, and demonstrate easily implemented approaches for mitigating different errors before a quantum computation experiment. On the investigated devices we find non-negligible qubit reset errors that cannot be parametrized as a diagonal mixed state, but manifest as a coherent phase of a superposition with a small contribution from the qubit's excited state. We are able to mitigate such errors by applying pre-rotations on the initialized qubits, which we demonstrate with multi-qubit entangled states. Our results demonstrate that Bayesian estimation can resolve small parameters - including those pertaining to quantum gate errors - with a high relative accuracy, at a lower measurement cost as compared with standard characterization approaches.
Highlights
Quantum computers experience a variety of errors that limit the ability of near-term quantum devices to perform arbitrary computations in the absence of fault-tolerant quantum error correction
In this subsection we present results from the selfconsistent estimation of the five state preparation and measurement (SPAM) parameters
We have not pursued this line of research, the available information about the parameter correlations and interdependency can be valuable for specific purposes
Summary
Quantum computers experience a variety of errors that limit the ability of near-term quantum devices to perform arbitrary computations in the absence of fault-tolerant quantum error correction. A variety of error mitigation techniques have been proposed to reduce the effects of errors in current devices, with the goal of achieving useful quantum computation on noisy near-term quantum computers [2,3] Many of these techniques focus on mitigating measurement errors through classical postprocessing of measurement data based on an accurately characterized model of errors in measurement [4–8]. Other commonly used methods for estimating average gate errors in gatesets are randomized benchmarking (RB) and many related randomized protocols [20–24] These decouple the effect of SPAM errors by fitting decay parameters to an exponential noise model as a function of sequence lengths and are used extensively in current experiments [25–30]. We estimate small gate errors and study the precision with which the Bayesian approach can be used to estimate and correct those, consistently finding a significant reduction in the required experiment measurements for the same level of precision as compared with standard characterization and error amplification techniques
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
