Machine learning for quantum control

Quantum control is valuable for quantum technologies such as high-fidelity quantum gates, adaptive quantum-enhanced metrology, and ultra-cold atom manipulation. Although both supervised and reinforcement learning are used to optimize control parameters in classical systems, quantum control for parameter optimization is mainly achieved via gradient-based greedy algorithms. However, greedy algorithms can yield poor results for quantum control, especially for highly constrained large-dimensional quantum systems. We employ differential evolution algorithms to circumvent the stagnation problem of non-convex optimization, and we average over the objective function to improve quantum control fidelity for noisy systems. To reduce computational cost, we introduce heuristics for early termination of runs and for adaptive selection of search subspaces. Our implementation is massively parallel and vectorized to reduce run time even further. We demonstrate our methods with two examples, namely quantum phase estimation and quantum gate design, for which we achieve superior fidelity and scalability than obtained using greedy algorithms.