Measuring and Suppressing Error Correlations in Quantum Circuits

Quantum error correction, fundamental to enabling large-scale quantum computing, relies on stochastic errors throughout a quantum circuit. Correlated errors between sequential logic gates violate this requirement, but are a realistic element of laboratory environments. To facilitate QEC it is necessary to identify and suppress such errors at both the physical and virtual layers of the quantum processor architecture.
We provide an analytic framework to identify correlated errors in randomly composed quantum circuits using only projective measurements at their conclusion. Using a single trapped ¹⁷¹Yb⁺ ion, we identify signatures of error correlations in the presence of engineered noise with tuneable correlation length. To reduce error correlations before the application of QEC, we work at a higher abstraction layer than the physical gates, replacing primitive qubit operations with logically equivalent dynamically corrected gates (DCGs) to form a virtual layer. We demonstrate that even in the presence of strongly correlated noise the signatures of error correlations at the virtual layer appear similar to standard gates exposed to uncorrelated noise, quantitatively extracting a >100x reduction in the correlated error component.