Quantum Computers Gain Robust Error Checks with New Gate Families

Researchers at the University of Edinburgh, in collaboration with National Quantum Computing and University of Warwick, have developed a family of scalable quantum accreditation protocols designed to rigorously define an upper bound on discrepancies between ideal and actual quantum calculations. The protocols address a critical challenge within the field of quantum computing, offering a practical and robust set of tools to assess the reliability of quantum circuits employing non-Clifford two-qubit gates. These gates, including families such as the fSim and XY gates, are increasingly native to existing and developing quantum hardware platforms, making this advancement particularly significant. Moreover, the team’s generalised Pauli twirling technique represents a potentially valuable advancement beyond existing error mitigation strategies, offering a novel approach to error characterisation and reduction.

Accreditation protocols now verify quantum circuits employing non-Clifford gates

Previously, a significant limitation in quantum computation involved the necessity to recompile quantum circuits to utilise only Clifford gates for verification purposes. This recompilation process could increase circuit depth by up to four times, introducing substantial overhead and hindering the practical application of quantum algorithms. The new accreditation protocols overcome this limitation by directly addressing circuits utilising non-Clifford gates. They establish a quantifiable upper bound on the total variation distance (TVD), a measure that rigorously quantifies the difference between the probability distributions of ideal, error-free quantum computations and those produced by an erroneous quantum computer. This is achieved without the need for extensive circuit rewriting or the introduction of significant computational overhead. The protocols are specifically designed for circuits utilising non-Clifford two-qubit gates, such as those from the fSim and XY families, which are prevalent in contemporary quantum hardware architectures.

This advancement allows for the first verification of analogue quantum simulations, a feat previously inaccessible without incurring prohibitive computational costs. Analogue quantum simulation, which leverages the natural dynamics of quantum systems to model other quantum systems, is a promising avenue for tackling complex scientific problems. However, verifying the accuracy of these simulations has been a major hurdle. Quantum accreditation protocols are now applicable to circuits utilising non-Clifford two-qubit gates, extending beyond systems limited to Clifford gates alone. The protocols accommodate a wide range of gate families including fSim, √iSWAP, the Sycamore gate, and Jaksch gates, demonstrating their versatility and broad applicability. Extending accreditation to these gate types unlocks the potential for more efficient and reliable quantum computation, enabling the development of more complex and powerful quantum algorithms. The protocols demonstrate robustness even with small perturbations to the system, maintaining their accuracy under realistic operating conditions, and accreditation is possible for both τ-decomposable and XY-decomposable gates under specific error models. Currently, however, these protocols assume a limited error model, primarily focusing on depolarising noise, and do not yet demonstrate accreditation across all noise types found in real-world quantum devices, such as amplitude damping or phase damping. Therefore, broadening the scope of applicable noise models is a key area for future research, alongside investigating the impact of correlated errors.

Validating quantum computations through maximum deviation benchmarks

As the field of quantum computing progresses beyond purely theoretical demonstrations and towards practical applications, establishing the accuracy of quantum calculations is paramount. The development of robust verification methods is crucial for building trust in quantum computers and ensuring the reliability of their results. These new protocols provide a method for quantifying how far a flawed quantum computer’s output deviates from the ideal result, representing an important step towards building dependable machines. Measuring the maximum possible difference, through upper-bounding total variation distance, provides a key benchmark for assessing the reliability of current quantum devices and establishes a baseline for refining techniques to pinpoint and correct specific errors within increasingly complex quantum circuits. The TVD, in this context, represents the largest possible difference between the probability of obtaining a specific outcome on an ideal quantum computer and the probability of obtaining the same outcome on a noisy quantum computer.

This accreditation process moves beyond reliance on simplified circuits, establishing a method for verifying quantum computations performed with current hardware. The protocols leverage techniques such as Pauli twirling, a process of averaging over a set of Pauli operators, to effectively mitigate the impact of certain types of noise and extract meaningful information about the underlying error characteristics. The generalised Pauli twirling technique employed by the researchers extends the applicability of this method to non-Clifford gates, which are essential for achieving quantum advantage. Consequently, a path opens towards quantifying the performance of increasingly complex quantum circuits and determining how to pinpoint the sources of error within them. This is a crucial step towards developing effective error correction strategies and building fault-tolerant quantum computers. However, it is important to note that the current protocols focus on identifying the maximum potential deviation, rather than providing a complete characterisation of the actual sources of error. Future research will likely focus on combining these accreditation protocols with techniques for error diagnosis and localisation, allowing for a more comprehensive understanding of the performance limitations of quantum devices and facilitating the development of targeted error mitigation strategies. The ability to accurately assess and improve the reliability of quantum computations is fundamental to realising the full potential of this transformative technology, with implications for fields ranging from materials science and drug discovery to financial modelling and artificial intelligence.

The researchers developed quantum accreditation protocols to assess the reliability of quantum circuits utilising complex, non-Clifford gates. These protocols establish an upper bound on the total variation distance, quantifying the maximum difference between ideal and noisy quantum computations. This matters because it provides a practical and scalable method for verifying computations on existing quantum hardware, moving beyond simpler circuit tests. The authors intend to combine these protocols with techniques for identifying and localising specific errors, further improving the accuracy of quantum computations.