We compute the expected randomized benchmarking sequence fidelity for a system subject to Gaussian time-correlated noise. For single qubit benchmarking we show that the expected sequence fidelity is given by the partition function of a long-range coupled spin-one Ising model, with each site in the Ising model corresponding to a free evolution interval. For d-state systems, the expected sequence fidelity is given by an Ising-like model partition function whose site variables are given by the weights of the adjoint representation of SU(d). A high effective temperature expansion for the partition function in the single qubit case shows decay of sequence fidelity varying from exponential for uncorrelated noise to a power law for quasistatic noise. Fitting an exponential to the sequence fidelity decay under correlated noise gives unreliable estimates of the average gate error rate.
We compute the expected randomized benchmarking sequence fidelity for a system subject to Gaussian time-correlated noise. For single qubit benchmarking we show that the expected sequence fidelity is given by the partition function of a long-range coupled spin-one Ising model, with each site in the Ising model corresponding to a free evolution interval. For d-state systems, the expected sequence fidelity is given by an Ising-like model partition function whose site variables are given by the weights of the adjoint representation of SU(d). A high effective temperature expansion for the partition function in the single qubit case shows decay of sequence fidelity varying from exponential for uncorrelated noise to a power law for quasistatic noise. Fitting an exponential to the sequence fidelity decay under correlated noise gives unreliable estimates of the average gate error rate.
Randomized benchmarking (RB) has emerged as a popular and easy-to-implement experimental technique for gauging the quality of gate operations in quantum computing devices. A typical randomized benchmarking procedure identifies the exponential decay in the fidelity as the benchmarking sequence of gates increases in length, and the decay rate is used to estimate the fidelity of the gate. Our current theory of RB, based on the assumption of time-independent or static noise in the gates, predicts an exponential decay in the fidelity. Observed deviations from the standard exponential decay are usually attributed to some amount of time correlations in the noise, though the precise mechanisms for deviation have yet to be fully explored. In this paper, we examine this question of randomized benchmarking for time-correlated noise---specifically for time-correlated dephasing noise for exact solvability---and elucidate the circumstances in which a deviation from exponential decay can be expected, as well as point out situations where exponential decay prevails despite the presence of time correlations.
In practical randomized benchmarking of logical gates for quantum computation, partial averaging is sometimes used to simplify the implementation. For instance, for two-qubit gates, averaging over only single-qubit operations is much easier to realize than the full averaging. We analyze such simplified, partial averaging and demonstrate that, unlike for the standard randomized benchmarking, the measured decay law of fidelity is a linear combination of several exponentials with different decay factors (3 for two qubits and single-bit averaging). The evolution with the sequence length is given by an iteration matrix, whose eigenvalues give the decay factors. For generic two-qubit gates two contributions decay fast and are barely visible, while the third factor characterizes a combination of gate errors in three channels. While the decay factor of this slowest exponential is close to that in the standard randomized benchmarking, we demonstrate that they differ in general and find this correction to the leading order. Furthermore, via a connection to the local invariants of two-qubit gates we find all exceptional gates, with more than one slow exponential, and analyze possibilities to extract their decay factors from the measured curves.
Randomized benchmarking (RB) protocols have become an essential tool for providing a meaningful partial characterization of experimental quantum operations. While the RB decay rate is known to enable estimates of the average fidelity of those operations under gate-independent Markovian noise, under gate-dependent noise this rate is more difficult to interpret rigorously. In this paper, we prove that single-qubit RB decay parameter p coincides with the decay parameter of the gate-set circuit fidelity , a novel figure of merit which characterizes the expected average fidelity over arbitrary circuits of operations from the gate-set. We also prove that, in the limit of high-fidelity single-qubit experiments, the possible alarming disconnect between the average gate fidelity and RB experimental results is simply explained by a basis mismatch between the gates and the state-preparation and measurement procedures, that is, to a unitary degree of freedom in labeling the Pauli matrices. Based on numerical evidence and physically motivated arguments, we conjecture that these results also hold for higher dimensions.
Randomized benchmarking (RB) is a widely adopted protocol for estimating the average gate fidelity in quantum hardware. However, its standard formulation relies on the assumption of temporally uncorrelated noise, an assumption often violated in current devices. In this work, we derive analytic expressions for the average sequence fidelity (ASF) in the presence of temporally correlated (non-Markovian) noise with classical memory, including cases where such correlations originate from interactions with a quantum environment. We show how the ASF can be interpreted to extract meaningful benchmarking parameters under such noise and identify classes of interaction Hamiltonians that render temporal correlations completely invisible to RB. We further provide operational criteria for witnessing temporal correlations due to quantum memory through RB experiments. Importantly, while classical correlations may remain undetectable in the ASF data, they can nonetheless significantly affect worst-case errors quantified by the diamond norm, a metric central to fault-tolerant quantum computing. In particular, we demonstrate that temporal correlations may suppress worst-case errors, highlighting that temporal correlations may not always have detrimental effects on gate performance.
In this Letter we propose a fully scalable randomized benchmarking protocol for quantum information processors. We prove that the protocol provides an efficient and reliable estimate of the average error-rate for a set operations (gates) under a very general noise model that allows for both time and gate-dependent errors. In particular we obtain a sequence of fitting models for the observable fidelity decay as a function of a (convergent) perturbative expansion of the gate errors about the mean error. We illustrate the protocol through numerical examples.
Randomized benchmarking (RB) protocols have become an essential tool for providing a meaningful partial characterization of experimental quantum operations. While the RB decay rate is known to enable estimates of the average fidelity of those operations under gate-independent Markovian noise, under gate-dependent noise this rate is more difficult to interpret rigorously. In this paper, we prove that single-qubit RB experiments provide direct estimates of the gateset circuit fidelity, a novel figure of merit which characterizes the expected average fidelity over arbitrary circuits of operations from the gateset. We also prove that, in the limit of high-fidelity single-qubit experiments, the difficulty introduced by gate-dependence reduces to a basis mismatch between the gates and the state-preparation and measurement procedures, that is, to a unitary degree of freedom in labeling the Pauli matrices. Based on numerical evidence and physically motivated arguments, we conjecture that these result hold for higher dimensions.