2020/03/04 by McKay, David C., Cross, Andrew W., Jay Gambetta +2 · 2 citations
Decision Sciences · Engineering · #FOS: Physical sciences #Manufacturing Process and Optimization #Quantum Physics (quant-ph) #Simulation Techniques and Applications
paper · pdf · doi:10.48550/arxiv.2003.02354
openalex publication_date 2020/03/04 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
To improve the performance of multi-qubit algorithms on quantum devices it is critical to have methods for characterizing non-local quantum errors such as crosstalk. To address this issue, we propose and test an extension to the analysis of simultaneous randomized benchmarking data -- correlated randomized benchmarking. We fit the decay of correlated polarizations to a composition of fixed-weight depolarizing maps to characterize the locality and weight of crosstalk errors. From these errors we introduce a crosstalk metric which indicates the distance to the closest map with only local errors. We demonstrate this technique experimentally with a four-qubit superconducting device and utilize correlated RB to validate crosstalk reduction when we implement an echo sequence.