We present experimental validation of a software-defined geometric noise suppression protocol on a commercial superconducting quantum processor (Rigetti Ankaa-3 via AWS Braket). The protocol combines three elements: (1) time-reversal symmetric Figure-8 control pulses that create closed trajectories in parameter space, (2) node-synchronized stroboscopic readout that gates measurements at loop-closure times, and (3) adaptive phase tracking that compensates for environmental drift. Key experimental results include: break-even threshold exceeded with +10 percentage point improvement over baseline idle; phase-dependent noise suppression achieving +82 percentage point improvement through optimal phase selection; and discovery of amplitude-phase coupling (APC) where optimal phase shifts by ≈ 0.9 rad per unit amplitude change. This coupling, interpretable as an environmentally-induced Berry connection in control parameter space, explains why conventional fixed-parameter calibration fails at varying drive strengths. The adaptive protocol successfully tracks environmental drift, recovering from 13% to 85% fidelity autonomously. These results demonstrate that geometric structure in quantum control can be exploited for robust, hardware-agnostic noise suppression without physical device modification.
We present experimental validation of a software-defined geometric noise suppression protocol on a commercial superconducting quantum processor (Rigetti Ankaa-3 via AWS Braket). The protocol combines three elements: (1) time-reversal symmetric Figure-8 control pulses that create closed trajectories in parameter space, (2) node-synchronized stroboscopic readout that gates measurements at loop-closure times, and (3) adaptive phase tracking that compensates for environmental drift. Key experimental results include: break-even threshold exceeded with +10 percentage point improvement over baseline idle; phase-dependent noise suppression achieving +82 percentage point improvement through optimal phase selection; and discovery of amplitude-phase coupling (APC) where optimal phase shifts by ≈ 0.9 rad per unit amplitude change. This coupling, interpretable as an environmentally-induced Berry connection in control parameter space, explains why conventional fixed-parameter calibration fails at varying drive strengths. The adaptive protocol successfully tracks environmental drift, recovering from 13% to 85% fidelity autonomously. These results demonstrate that geometric structure in quantum control can be exploited for robust, hardware-agnostic noise suppression without physical device modification.
Fault-tolerant quantum computing requires high-fidelity quantum gates that are resilient to control errors and device imperfections. Here, we develop energy-optimal control pulses for single-qubit gates that are simultaneously robust against frequency detuning and waveform inhomogeneities while also suppressing leakage outside the computational subspace. Combining the optimized pulse envelopes with an adaptive control scheme, we experimentally demonstrate optimal robust gates on a superconducting quantum processor. For a gate duration of 90ns, we achieve fidelities above 99.9% for the X gate and 99.6% for Hadamard-type gates. For the X gate, we maintain fidelities above 99.5% for simultaneous drive-amplitude variations up to 20% and qubit detuning up to 4 MHz. This robustness outperforms that of conventional Gaussian pulses by an order of magnitude, without much difference in peak gate fidelity. These results establish a route towards robust single-qubit control in large-scale quantum processors, advancing the development of fault-tolerant quantum computing.
Noise characterisation not only enables error mitigation but also provides valuable information about the intrinsic performance of a quantum device, identifying dominant error contributions and bounding noise stability well beyond the aggregated calibration data. Deploying such methods on a specific hardware topology must combine efficiency and accuracy, assess noise stability, and validate the reconstructed noise model. Here we develop and implement these methods on a superconducting processor with square-lattice topology. First, we revise the cycle-benchmarking protocol and reduce the bias arising from unlearnable degrees of freedom via a regularised cost function. Second, we extend a previously developed tensor-network tomography method to match the device layout and show how its accuracy can be improved by repeated layer execution. Both methods allow approximate sublayer noise models to be extracted from parent ones. Using noise characterisation as a tool, we map the noise structure, track drift, and validate the adequacy of the learnt noise models against experimentally observed noisy estimations in benchmark physical and chemistry circuits. These results inform further improvements in the underlying technology and provide estimates of the size of noise-mitigated circuits that the device can support.
We report phase-resolved measurements of amplitude-phase coupling (APC) in strongly driven superconducting transmon qubits, a symmetry-based mechanism analysis, and follow-up experiments on a second device. On a Rigetti Ankaa-3 transmon (1000 shots per point, five drive amplitudes, matched root-mean-square drive), the fitted optimal control phase
ϕ∗(A) drifts by more than 3 rad across the explored amplitude range for both a rectangular reference pulse and a self-intersecting Figure-8 trajectory, and the phase contrast of the two waveform families exhibits a reproducible low-/high-amplitude crossover. Theoretically, we prove that in the resonant rotating frame of the three-level Duffing model a constant-phase rectangular drive cannot produce any phase-dependent population response:
CRect=0 exactly. Any measured rectangular phase contrast is therefore a probe of physics outside this model (beyond-rotating-wave carrier-phase dependence, control-chain distortion, readout imperfections). The Figure-8 waveform breaks the symmetry intrinsically through its 2:1 Fourier content; a third-order resonance
ω+ω−2ω=0 mediated by the anharmonic level
∣2⟩ predicts
CF8∝A6N2 at weak drive, confirmed by direct numerical propagation (fitted log-log slope
5.97±0.02; loop-number ratio 2434 vs. predicted 2500). On the successor device Cepheus-1-108Q, the predicted phase contrast lies below the binomial shot-noise floor at the accessible amplitudes, but the same dataset reveals a stronger signature of the underlying first-order average-Hamiltonian cancellation: the phase-averaged survival probability of the Figure-8 train remains pinned at
⟨P(∣0⟩)⟩ϕ≈0.985 across all tested conditions, while the isenergetic rectangular reference sweeps the full range 0.24-0.93—an approximately 70-fold suppression of the Rabi-drive response. Together these results identify APC as a measurable control-landscape feature with distinct intrinsic and extrinsic components, and demonstrate a calibration-free, amplitude-insensitive idling property of Figure-8 pulse trains.
We report phase-resolved measurements revealingamplitude–phase coupling (APC) in a strongly drivensuperconducting qubit. Using 1000 shots per point acrossfive drive amplitudes on a Rigetti Ankaa-3 transmonqubit (Qubit 58; T1 = 36.7 μs, T2 = 22.2 μs), we comparea rectangular reference pulse and a self-intersectingFigure-8 trajectory under matched root-mean-squaredrive amplitude. From harmonic fits to the phasedependentsurvival probability, we extract the optimalcontrol phase ϕ∗(A) and show that it shifts by more than3 rad across the explored amplitude range for both pulsefamilies. To quantify phase sensitivity, we define thephase contrast C(A) = maxϕ ˆ F −minϕ ˆ F and the contrastratio R(A) = CF8/CRect. At low amplitude (A ≤ 0.6),the Figure-8 trajectory reduces phase contrast relativeto the rectangular reference (R = 0.48–0.57), whereas athigh amplitude (A ≥ 0.8) it increases it (R = 1.22–1.45),revealing a crossover in trajectory-dependent phase sensitivity.An independent replication acquired in a separatecalibration session at 100 shots per point preserves thesame low- and high-amplitude sign structure. Additionalpulse-train continuity tests show that internal timinggaps reduce phase contrast without measurably shiftingthe optimal phase, and that a fixed 4 ns gap showsno statistically significant dependence on its placementwithin the train. These observations identify APC as anexperimentally accessible feature of the measured controllandscape and show that fixed-phase calibration canbecome systematically biased across drive amplitudes inthe strongly driven regime.