Efficient pulse-sequence optimization method for quantum control under realistic hardware constraints

We propose a simple yet efficient pulse-sequence optimization method for controlling quantum systems. The ansatz leads to an efficient exploration of the unitary space without over-parameterization, even when the pulse amplitudes are restricted to a few discrete values suitable for quantum devices. Bandwidth and power limitations of experimental settings can be straightforwardly included with only minor computational overhead. We numerically validate the method by applying it to (i) unitary synthesis of a three-qubit gate and (ii) ground-state preparation on a globally-driven Rydberg-atom platform, and (iii) state transfer in an Ising spin chain. For all problems considered, our method approaches an information-theoretic lower bound on the number of parameters and exhibits advantages when compared to commonly used quantum control algorithms. For example, our method achieves higher accuracy with substantially reduced computational overhead, requiring fewer figure-of-merit evaluations and shorter simulation runtimes, even under realistic hardware constraints.