Variational Quantum Optimization of Nonlocality in Noisy Quantum Networks

We consider the nonlocal correlations accessible to noisy quantum communication networks. We focus on chain and star topologies with entanglement linking adjacent nodes. To find maximally nonlocal correlations, we apply variational quantum optimization across a wide range of network configurations, resources, and noise models. In this procedure, parameterized quantum circuits efficiently simulate noisy quantum networks on a quantum computer. Nonlocality is quantified using known Bell inequalities. Gradient ascent is used to maximize the nonlocality while gradients are computed using numerical techniques compatible with quantum hardware. Using our optimization framework, we investigate the effect of different noise models on the nonlocal correlations of quantum networks with fewer than 20 qubits. We analyze our numerical results to evaluate the noise robustness of network nonlocality. Our work helps inform the development of near-term quantum communication protocols and networks. Furthermore, as larger quantum computers become available, our optimization technique shows promise in scaling beyond the limits of classical computing .