Optimizing performance of quantum operations with non-Markovian decoherence: the tortoise or the hare?

The interaction between a quantum system and its environment limits our ability to control it and perform quantum operations on it. We present an efficient method to find optimal controls for quantum systems coupled to non-Markovian environments, by using the process tensor to compute the gradient of an objective function. We consider state transfer for a driven two-level system coupled to a bosonic environment, and characterize performance in terms of speed and fidelity. We thus determine the best achievable fidelity as a function of process duration. We show there is a trade-off between speed and fidelity, and that slower processes can have higher fidelity by exploiting non-Markovian effects.