Wu and Colleagues Introduces Cyclic Control Strategy for Fast CZ Gate Fidelity

A new cyclic control strategy overcomes the trade-off between gate speed and accuracy in quantum computing. Ze-An Zhao and colleagues expand the parameter space for pulse design, enabling strong suppression of coherent errors in a superconducting controlled-Z gate. The average error reduces from 0.27% to 0.12% without extending gate duration. This advancement provides a general pathway towards achieving both fast and high-fidelity quantum gates, representing a key step towards scalable quantum computation.

Restored qubit controllability enables high-fidelity, fast superconducting controlled-Z gates

Error rates dropped to 0.12%, a significant reduction from 0.27% in superconducting controlled-Z gates, representing a major leap in quantum gate fidelity. The team at University of Science and Technology of China achieved this improvement without increasing gate duration, surpassing the conventional speed-fidelity trade-off which previously demanded slower gate speeds for higher accuracy. By addressing distortions in control pulses, tiny imperfections disrupting precise qubit operation, they expanded the range of adjustable parameters during gate operation, effectively restoring controllability.

A novel cyclic control strategy provides a general pathway towards building faster and more reliable quantum computers, circumventing a key limitation in current superconducting quantum circuit designs. Validation of the approach used cross-entropy benchmarking, a method for assessing quantum gate performance by measuring preservation of quantum information. This revealed a reduction in coherent errors, stemming from imperfections within the quantum system, and successful suppression of leakage, unwanted transitions to unintended quantum states, alongside phase errors, all without extending gate operation duration, a critical step towards practical quantum computation. The team discovered that short-term distortions in control pulses disrupt the time-reflection symmetry of waveforms, increasing constraints on precise qubit operation, and overcame this by adding an extra adjustable parameter to the control process.

Waveform optimisation reduces coherent errors in a superconducting controlled-Z gate

Implementation of a parameter-space expansion technique on a superconducting controlled-Z gate successfully restored controllability by adding a degree of freedom to the control pulses. The authors acknowledge, however, that their demonstration is limited to this specific gate type, and broader applicability to other quantum gates or different quantum computing platforms remains unproven, necessitating further investigation to establish wider utility. Conventional methods, such as smoothing pulse envelopes or lengthening pulse duration, were cited as approaches sacrificing speed to improve fidelity.

Distortions in control pulses over short timescales break the time-reflection symmetry of waveforms, increasing the constraints on precise gate operation. The authors do not address potential caveats regarding scalability of this approach to larger quantum circuits or more complex gate sequences. While the reduction in coherent error is striking, the impact on overall circuit performance requires further evaluation using more intricate quantum algorithms. Furthermore, the authors do not discuss limitations related to the specific hardware used in their experiment or the challenges of implementing this control strategy on different superconducting qubit designs.

Coherent error reduction via expanded parameter control in superconducting qubits

Researchers at University of Science and Technology of China have developed a new cyclic control strategy that reduces coherent errors in quantum gates without extending gate duration. A reduction in average coherent error from 0.27% to 0.12% was achieved across multiple two-qubit gates, a result verified using cross-entropy benchmarking. This improvement addresses a long-standing challenge in quantum computing: the trade-off between the speed and accuracy of quantum operations.

The demonstrated cyclic control strategy was implemented on a superconducting controlled-Z gate, a common type of two-qubit gate used in quantum processors. This work builds upon recent advances in superconducting quantum computing, bringing the field closer to achieving quantum advantage, demonstrating a quantum computer’s ability to solve problems beyond the reach of classical computers. Progress in reducing logical error rates and utilising higher-rate qLPDC codes are cited as examples of this momentum.

Two-qubit gates continue to represent a key bottleneck, exhibiting sharply higher error rates than single-qubit operations. The team notes that hardware imperfections disrupt the cyclic evolution essential for many quantum gates, including Mølmer-Sørensen gates and geometric phase gates. Rapid fluctuations in control signals pose a particular challenge, although longer timescale distortions can be mitigated through calibration and filtering. This approach offers a new avenue for quantum gate design, circumventing limitations imposed by waveform distortions that typically force a trade-off between speed and accuracy. Expanding the available control parameters restored controllability over the quantum system, enabling precise manipulation of qubits without extending operation duration. This parameter-space expansion addresses distortions affecting pulse accuracy, a common issue in superconducting circuits, and successfully suppressed coherent errors, tiny imperfections causing information loss, in controlled-Z gates.

The researchers successfully demonstrated a new cyclic control strategy that reduced the average coherent error in two-qubit superconducting controlled-Z gates from 0.27% to 0.12%. This achievement overcomes a persistent trade-off between the speed and accuracy of quantum operations, a significant hurdle in building scalable quantum computers. By expanding the available control parameters, the team restored controllability over the quantum system and suppressed errors without increasing gate duration. The findings, validated by cross-entropy benchmarking, present a general route to faster, more reliable quantum gates.