Superconducting Qubit Gates Achieve 15 Times More Robustness to Parameter Fluctuations

Maintaining the delicate quantum states necessary for practical quantum computation presents a significant challenge, as even small fluctuations in a qubit’s operating parameters can introduce errors. Emily Wright, Leo Van Damme, and Niklas J. Glaser, at the Technical University of Munich, alongside colleagues including Amit Devra and Federico A. Roy, now demonstrate a method for creating superconducting qubit gates that actively resist these disturbances. The team developed specially engineered pulses, using a technique called gradient ascent pulse engineering, which dramatically improve gate performance even when faced with variations in frequency, drive amplitude, and coherence. These robust gates not only suppress errors caused by gradual drifts in operating conditions, but also exhibit resilience to random, time-dependent noise, representing a substantial step towards building more stable and reliable quantum computers.

State-of-the-art single-qubit gates on superconducting qubits achieve the performance needed for error-corrected computations, but maintaining this level of fidelity is challenging due to qubit instabilities, environmental changes, and control inaccuracies. These fluctuations introduce errors that limit the coherence and reliability of quantum computations, necessitating robust calibration and control techniques. This research addresses this challenge by developing a method to characterise and mitigate these parameter fluctuations, improving the stability and accuracy of quantum operations and enhancing the feasibility of large-scale quantum computing.

Optimized Control Pulse Generation and Calibration

This document details the methods used to implement and calibrate robust quantum gates, specifically Xπ/2 gates, on a superconducting qubit. The approach focuses on creating gates that are less susceptible to errors and more reliable in real-world conditions. The core method used to design the control pulses is GRAPE, a numerical optimization technique that finds the pulse shape that best achieves the desired quantum gate operation. Two main pulse shapes were explored: FROG, a standard pulse shaping technique, and AFROG, an improved version of FROG designed to be more robust to control errors. The optimization process involved minimizing a cost function and calculating gradients to refine the pulse shapes, determining specific Fourier coefficients for each pulse, allowing for precise reconstruction of the control signals.

The calibration process was performed sequentially, tuning each parameter individually. A Ramsey sequence was used to determine the qubit frequency, and the pulse amplitude was adjusted to achieve the desired population. Error amplification sequences were employed to fine-tune the gate parameters, and the DRAG technique was used to suppress unwanted transitions and further improve gate fidelity.

Robust Quantum Gates Suppress Parameter Fluctuations

Scientists have developed robust quantum gates that significantly suppress errors caused by fluctuations in qubit parameters. The research demonstrates a new approach to gate design, achieving substantial improvements in performance compared to standard techniques. These gates are specifically engineered to be resilient to both static and time-dependent noise, a critical challenge in building stable quantum computers. The team designed pulses to implement robust Xπ/2 gates, accounting for potential leakage effects by including the lowest three energy levels of the qubit system. Simulations utilized a time step corresponding to the sampling rate of the arbitrary waveform generator.

An advanced AFROG pulse, optimized for both detuning and amplitude error, employed a higher number of time steps and a wider range of detuning. Measurements confirm that the robust pulses suppress coherent errors from drive amplitude drifts more effectively than standard techniques with derivative removal by adiabatic gate corrections. Furthermore, these robust gates demonstrate resilience to stochastic, time-dependent noise, suppressing added errors during increases in dephasing more effectively than standard corrections. The team meticulously calibrated each pulse parameter, first characterizing the qubit frequency using a Ramsey pulse sequence. Error amplification sequences were employed to fine-tune the amplitude, frequency, and DRAG correction parameter, demonstrating the precision of the calibration process. These results represent a significant step towards building more reliable and stable quantum computers capable of performing complex calculations.

Robust Gates Resist Noise and Drift

This research demonstrates the effectiveness of newly designed single-qubit gates in maintaining stable operations despite variations in key parameters and the presence of noise. Scientists developed the Frequency RObust Gate (FROG) and the Amplitude-and-frequency RObust Gate (AROG) using a gradient-based numerical optimization method, creating gates that exhibit resilience to both static and time-dependent errors. The team found that AROG gates significantly outperform conventional techniques in the presence of amplitude drifts, experiencing less added error during fluctuations. Furthermore, both FROG and AROG gates demonstrate improved resilience to dephasing, suppressing added errors compared to standard corrections.

These improvements are particularly relevant for fault-tolerant quantum computing, as the narrower distribution of sequence fidelity observed with AROG gates suggests a lower worst-case error rate. The researchers connected observed amplitude deviations to temperature variations in the control hardware, highlighting the potential for these robust gates to maintain operation without recalibration. While acknowledging that further advances in shielding and fabrication continue to address noise sources, this work confirms that robust gate designs also offer resilience to stochastic errors, such as those caused by fast frequency fluctuations.