Fast, noise-robust pulses for parametric entangling gates in superconducting qubits

For Noisy Intermediate-Scale Quantum (NISQ) devices, incorporating robustness into computing operations is a critical target for enhancing computational capability. Superconducting transmon qubits are a world-leading quantum computing platform, which incorporate both active and passive strategies to improve performance, such as operating in flux-control “sweet-spots” for entangling parametric gates. We demonstrate that, even when starting from a set of parameters outside the sweet-spot condition, the introduction of numerically optimized flux-modulation waveforms can restore gate robustness. Our approach involves specifiable deviation from the sinusoidal flux waveforms conventionally applied in parametric gates, up to the limit of arbitrary waveform optimization. This modulation enables the realization of faster gates while retaining the flux-noise robustness provided by dynamical sweet spots. We also demonstrate that this approach provides additional robustness to ambient dephasing noise by combining the modulated flux drive with optimized single-qubit drives applied concurrently. Finally, we show initial experimental results from the application of robust entangling gates to flux-tunable transmon qubits, validating the performance of this approach to gate construction.