Caleb Walton

The fidelity of a quantum gate is sensitive to small deviations in the physical control parameters. Unfortunately, it is generally difficult to exactly model the implemented Hamiltonian for a set of user-defined parameters, necessitating on-device calibration. Here, we present an active learning framework based on Bayesian optimization with a Gaussian Process surrogate to find the optimal parameter set. We validate the technique through numerical calibration of the laser amplitude and frequencies that implement the trapped-ion Mølmer Sørensen gate. We show that a Gaussian process can model the Hamiltonian dynamics. The addition of active learning accelerates the discovery of the optimal parameter set with speed and final fidelity dependent on the quantum projection noise of the data. These results establish the utility of active learning and surrogate models for quantum calibration and control.

Optical crosstalk due to imperfect addressing in trapped-ion entangling gates generates unwanted non-local entanglement between target ions and their neighbors that is difficult to mitigate using standard quantum error correction. We present a method to design entangling operations that are inherently insensitive to crosstalk by engineering the effective qubit coupling matrix. By controlling the geometric phases generated in the motional modes of the ion string, we construct a coupling matrix that selectively excludes crosstalk-affected neighbor ions from the entangling operation. This approach requires no knowledge of the amount of crosstalk present and avoids the need for additional gate operations or modifications to the optical setup. We numerically demonstrate the construction of crosstalk-insensitive entangling pulses for target ion pairs within an equispaced 20-ion string and provide experimental validation of crosstalk-insensitive entanglement in a three-ion string.