Model-free Distortion Canceling and Control of Quantum Devices

Quantum devices need precise control to achieve their full capability. In this work, we address the problem of controlling closed quantum systems, tackling two main issues. First, in practice the control signals are usually subject to unknown classical distortions that could arise from the device fabrication, material properties and/or instruments generating those signals. Second, in most cases modeling the system is very difficult or not even viable due to uncertainties in the relations between some variables and inaccessibility to some measurements inside the system. In this paper, we introduce a general model-free control approach based on deep reinforcement learning (DRL), that can work for any closed quantum system. We train a deep neural network (NN), using the REINFORCE policy gradient algorithm to control the state probability distribution of a closed quantum system as it evolves, and drive it to different target distributions. We present a novel controller architecture that comprises multiple NNs. This enables accommodating as many different target state distributions as desired, without increasing the complexity of the NN or its training process. The used DRL algorithm works whether the control problem can be modeled as a Markov decision process (MDP) or a partially observed MDP. Our method is valid whether the control signals are discrete- or continuous-valued. We verified our method through numerical simulations based on a photonic waveguide array chip. We trained a controller to generate sequences of different target output distributions of the chip with fidelity higher than 99%, where the controller showed superior performance in canceling the classical signal distortions.

Noise Detection with Spectator Qubits and Quantum Feature Engineering

Designing optimal control pulses that drive a noisy qubit to a target state is a challenging and crucial task for quantum engineering. In a situation where the properties of the quantum noise affecting the system are dynamic, a periodic characterization procedure is essential to ensure the models are updated. As a result, the operation of the qubit is disrupted frequently. In this paper, we propose a protocol that addresses this challenge by making use of a spectator qubit to monitor the noise in real-time. We develop a quantum machine-learning-based quantum feature engineering approach for designing the protocol. The complexity of the protocol is front-loaded in a characterization phase, which allow real-time execution during the quantum computations. We present the results of numerical simulations that showcase the favorable performance of the protocol.