Random circuit sampling is a leading approach to demonstrating quantum computational advantage, but benchmarking noisy implementations through linear cross-entropy requires costly calculations of ideal output probabilities. We propose a metrological benchmark based on the response to controlled perturbations, characterized by quantum Fisher information (QFI). A relation between QFI and out-of-time-order correlators enables protocols with local or global control. For Haar-random circuits, the average QFI approaches its maximal value in the local protocol and grows linearly with the number of qubits under collective control. In contrast, Clifford circuits yield zero QFI despite extensive operator spreading, showing that the response probes dynamical properties beyond spreading alone. A butterfly protocol further uses system size to enhance sensitivity, yielding a mean inverse sensitivity proportional to the number of qubits with single-qubit readout. Its fluctuations distinguish the Haar and Clifford ensembles despite their identical mean responses. For noisy implementations, we derive an exact relation between the noisy and ideal QFI within a global white-noise model, providing a quantitative reference for the degradation of the benchmark. These protocols enable metrological benchmarking of noisy random quantum circuits without computing ideal output probabilities.
AI Summary
The paper asks whether random quantum circuits can be benchmarked through controlled dynamical responses without computing ideal output probabilities, as required by linear cross-entropy benchmarking. It proposes preparing probes using a circuit $U$, an intervening operation, and $U^\dagger$, then measuring their response to a parameterized rotation. For pure probes, quantum Fisher information (QFI) is expressed through out-of-time-order correlators (OTOCs), including a squared OTOC that accesses fourth-order unitary statistics.
Two protocols use local or collective control. For $n$ qubits and $r$ repeated encoding rotations, Haar-averaged QFI approaches $r^2$ under local control and $nr^2$ under collective control. Both protocols give exactly zero QFI for Clifford circuits, despite their operator spreading. A butterfly variant uses collective encoding and single-qubit readout: its weak-signal mean inverse sensitivity approaches $n/2$ for both ensembles, while circuit-to-circuit variance approaches $n/(4\cdot2^n)$ for Haar circuits and $n/4$ for Clifford circuits. These results follow from Haar-moment calculations and Pauli-string statistics, with finite-size expressions supplied.
For a globally depolarized probe with parameter-independent ideal-state weight $f$, the exact relation is $I_{\mathrm{noisy}}=[f^2/(f+2(1-f)/2^n)]I_{\mathrm{ideal}}$. The single-qubit readout instead has approximately $f^2$ Fisher-information suppression near zero signal. The analysis requires inverse-circuit access; butterfly sensitivity formulas use weak-signal and large-system approximations. Noise is treated only through global white noise, with common $f$ assumed when rescaling ensemble fluctuations. Finite-depth circuits and realistic gate-dependent noise are identified as extensions needed for experimental application.
Little sign of AI-generated textStrong assumptionsMissing comparisonUnclear novelty
Error found: The butterfly protocol's claimed equality of Haar and Clifford mean inverse sensitivities is only an approximation: for one qubit at zero signal, the actual means are π/4 and 2/3, respectively.
For the stated butterfly circuit, the displayed readout expansion omits a generally nonzero zero-signal intercept and has the opposite slope sign, although its slope magnitude remains correct.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.
Random circuit sampling is a leading approach to demonstrating quantum computational advantage, but benchmarking noisy implementations through linear cross-entropy requires costly calculations of ideal output probabilities. We propose a metrological benchmark based on the response to controlled perturbations, characterized by quantum Fisher information (QFI). A relation between QFI and out-of-time-order correlators enables protocols with local or global control. For Haar-random circuits, the average QFI approaches its maximal value in the local protocol and grows linearly with the number of qubits under collective control. In contrast, Clifford circuits yield zero QFI despite extensive operator spreading, showing that the response probes dynamical properties beyond spreading alone. A butterfly protocol further uses system size to enhance sensitivity, yielding a mean inverse sensitivity proportional to the number of qubits with single-qubit readout. Its fluctuations distinguish the Haar and Clifford ensembles despite their identical mean responses. For noisy implementations, we derive an exact relation between the noisy and ideal QFI within a global white-noise model, providing a quantitative reference for the degradation of the benchmark. These protocols enable metrological benchmarking of noisy random quantum circuits without computing ideal output probabilities.
The paper asks whether random quantum circuits can be benchmarked through controlled dynamical responses without computing ideal output probabilities, as required by linear cross-entropy benchmarking. It proposes preparing probes using a circuit U, an intervening operation, and U†, then measuring their response to a parameterized rotation. For pure probes, quantum Fisher information (QFI) is expressed through out-of-time-order correlators (OTOCs), including a squared OTOC that accesses fourth-order unitary statistics.
Two protocols use local or collective control. For n qubits and r repeated encoding rotations, Haar-averaged QFI approaches r2 under local control and nr2 under collective control. Both protocols give exactly zero QFI for Clifford circuits, despite their operator spreading. A butterfly variant uses collective encoding and single-qubit readout: its weak-signal mean inverse sensitivity approaches n/2 for both ensembles, while circuit-to-circuit variance approaches n/(4⋅2n) for Haar circuits and n/4 for Clifford circuits. These results follow from Haar-moment calculations and Pauli-string statistics, with finite-size expressions supplied.
For a globally depolarized probe with parameter-independent ideal-state weight f, the exact relation is Inoisy=[f2/(f+2(1−f)/2n)]Iideal. The single-qubit readout instead has approximately f2 Fisher-information suppression near zero signal. The analysis requires inverse-circuit access; butterfly sensitivity formulas use weak-signal and large-system approximations. Noise is treated only through global white noise, with common f assumed when rescaling ensemble fluctuations. Finite-depth circuits and realistic gate-dependent noise are identified as extensions needed for experimental application.
Abstract We study the distribution over measurement outcomes of noisy random quantum circuits in the regime of low fidelity, which corresponds to the setting where the computation experiences at least one gate-level error with probability close to one. We model noise by adding a pair of weak, unital, single-qubit noise channels after each two-qubit gate, and we show that for typical random circuit instances, correlations between the noisy output distribution pnoisy p noisy and the corresponding noiseless output distribution pideal p ideal shrink exponentially with the expected number of gate-level errors. Specifically, the linear cross-entropy benchmark F that measures this correlation behaves as F=exp(−2sϵ±O(sϵ2)) F = exp ( - 2 s ϵ ± O ( s ϵ 2 ) ) , where ϵ ϵ is the probability of error per circuit location and s is the number of two-qubit gates. Furthermore, if the noise is incoherent—for example, depolarizing or dephasing noise—the total variation distance between the noisy output distribution pnoisy p noisy and the uniform distribution punif p unif decays at precisely the same rate. Consequently, the noisy output distribution can be approximated as pnoisy≈Fpideal+(1−F)punif p noisy ≈ F p ideal + ( 1 - F ) p unif . In other words, although at least one local error occurs with probability 1−F 1 - F , the errors are scrambled by the random quantum circuit and can be treated as global white noise, contributing completely uniform output. Importantly, we upper bound the average total variation error in this approximation by O(Fϵs) O ( F ϵ s ) . Thus, the “white-noise approximation” is meaningful when ϵs≪1 ϵ s ≪ 1 , a quadratically weaker condition than the ϵs≪1 ϵ s ≪ 1
Random circuit sampling (RCS) remains one of the most competitive frameworks for demonstrating quantum advantage in near-term noisy intermediate-scale quantum (NISQ) hardware. Unfortunately, absent error-correction, existing benchmarks to characterize these experiments, like linear cross-entropy, have been classically spoofed due to noise. Because of this, there are interesting regimes, like shallow-depth random quantum circuits, where sampling is plausibly classically intractable, but no existing benchmark can distinguish between a noisy quantum computer and an adversarial classical spoofer. In this paper, we demonstrate that the nonlinear cross-entropy provides a sample-efficient benchmark for shallow-depth all-to-all random quantum circuits whose score cleanly separates noisy quantum computers from state-of-the-art classical spoofers, even in the presence of depolarizing noise. Further, we develop a binary classifier based on the notion of heavy output generation that features logarithmic sample complexity at short depth. Our evidence comes from exact analytic expressions for all-to-all Brownian circuit ensembles derived using replica tricks, and numerical simulations that corroborate these results for discrete Haar-random unitary circuits.
Quantum technologies work by utilizing properties inherent in quantum systems such as quantum coherence and quantum entanglement and are expected to be superior to classical counterparts for solving certain problems in science and engineering. The quantum technologies are, however, fragile against an interaction with an environment (decoherence) and in order to utilize them with high accuracy we need to develop error mitigation techniques which reduce decoherence effects. In this work, we analyze quantum error mitigation (QEM) protocol for quantum metrology in the presence of quantum noise. We demonstrate the effectiveness of our QEM protocol by analyzing three types of quantum Fisher information (QFI), ideal (error-free) QFI, noisy (erroneous) QFI, and quantum-error-mitigated QFI, and show both analytically and numerically that the scaling behaviors of quantum-error-mitigated QFI with respect to the number of probes become restored to the those exhibited in the ideal quantum metrology. Our QEM protocol is constructed by an ensemble of quantum circuits, namely QEM circuit groups, and has advantages such that it can be applied to noisy quantum metrology for any type of initial state as well as any type of the probe-system Hamiltonian, and it can be physically implemented in any type of quantum device. Furthermore, the quantum-error-mitigated QFI become approximately equal to the ideal QFI for almost any values of physical quantities to be sensed. Our protocol enables us to use quantum entanglement as a resource to perform high-sensitive quantum metrology even under the influence of quantum noise.
We study quantum sensing with Floquet chaotic dynamics generated by Haar random unitary gates. The metrological resources consist of three ingredients: A given initial state, a set number of Haar random unitary gates and the sensing gates. There are two natural ways of organizing the resources: the first one is the "control" protocol, where the random unitary gates act as random controls and intertwine with the deterministic sensing gates and the second one is the "state-preparation" protocol, where random unitary gates play the role of preparing the metrological useful states. In each protocol, we consider both global Haar random unitary gates and a set of local two-site Haar random unitary gates that forms a Floquet random quantum circuit (RQC) respectively. We find linear, shot-noise scaling of the metrological precision, quantified by the quantum Fisher information (QFI), in the asymptotic limit when the Hilbert space dimension becomes large, and quantum advantages beyond linear scaling in the non-asymptotic regimes. We also bound the fluctuation of the QFI using concentration inequalities. Our analytical findings are corroborated by numerical simulations. Finally, along the way of analyzing the precision limit, we prove an empirical conjecture of RQC: In the asymptotic limit of large local Hilbert space dimension, the Floquet operator of a Floquet RQC essentially behaves like a global unitary operator.
Cross-entropy benchmarking is a central technique adopted to certify a quantum chip in recent investigations. To better understand its mathematical foundation and develop new benchmarking schemes, the concept of ergodicity was introduced to random circuit sampling and it was found that the Haar random quantum circuit could satisfy an ergodicity condition—the average of certain types of postprocessing function over the output bit strings is close to the average over the unitary ensemble. For noiseless random circuits, it was proven that the ergodicity holds for polynomials of degree t with positive coefficients when the random circuits form a unitary 2 t -design. For strong enough noise, the ergodicity condition is violated, which suggests that ergodicity is a property that can be exploited to certify a quantum chip. The deviation of ergodicity was formulated as a measure for quantum chip benchmarking, and it was demonstrated that it can be used to estimate the circuit fidelity for global depolarizing noise and weakly correlated noise. For a quadratic postprocessing function, our framework recovered Google's result on estimating the circuit fidelity via linear cross-entropy benchmarking (XEB), and we gave a sufficient condition on the noise model characterizing when such estimation is valid. The results establish an interesting connection between ergodicity and noise in random circuits and provide new insights into designing quantum benchmarking schemes.
Random quantum circuits aim to efficiently reproduce the statistical properties of ideal random quantum evolution. One approach is to construct approximate unitary designs, which match the moments of Haar-random unitaries up to a prescribed order with controlled error. We establish quantitative guarantees for how accurately these designs reproduce the distribution of each output probability. We show that, for every strong ε-approximate unitary k-design on n qubits, the distribution of each output probability is within O(k−1/2(1+k/2n)−1/2+ε) in Kolmogorov distance of the Porter-Thomas distribution, a beta distribution with parameters 1 and 2n−1. This bound is optimal up to constant factors, implying that a substantially better uniform bound requires additional structure. We then show that a mild form of local invariance suffices. We prove that any such design that is invariant under unitary translations acting on a single qubit approximates Porter-Thomas to total variation distance O(k−1/2+ε). Every strong ε-approximate unitary k-design invariant on logk+O(1) qubits achieves error 2−Ω(k)+O(ε) in Kolmogorov distance and 2−Ω(k)+O(εlog(2/ε)) in total variation distance. These results identify the distributional accuracy guaranteed by moment matching and the structure that improves it. As an application, we show that the Shannon entropy of the output distribution of a strong ε-approximate k-design is concentrated around its Haar value n−Θ(1): its mean differs from the Haar mean by O(k−1/4+ε1/2) and its standard deviation is O(2−n/2+k−1/8+ε1/4). Consequently, a trusted device sampling such circuits generates n−O(1) bits of entropy per sample, which makes them useful for random number generation.