QGASS: a gradient-free stochastic feedback control approach for arbitrary open quantum systems

High-fidelity state preparation represents a fundamental challenge in a variety of quantum technology applications. While the majority of optimal control approaches use feedback to improve the controller, the controller itself often does not incorporate explicit state dependence. Here, we present a system-agnostic approach for measurement-based feedback control of the stochastic master equation with optimality in mind. Our approach takes the approach of treating optimal control as a learning problem, and we train deep policy networks to serve as our controllers. The explicit feedback nature allows a variety of system and control structures that are prohibitive by many other techniques and can in effect react to unmodeled effects through nonlinear filtering. Our approach benefits from characteristics of both stochastic sampling and gradient-based optimization methods yet does not require differentiability as in backpropagation approaches. We demonstrate that this method is efficient due to inherent parallelizability, robust to open system interactions, and outperforms landmark state-dependent feedback control results in simulation. Finally, we show that the resulting algorithm can be applied in a plug-and-play manner to a variety of open quantum systems.