Learning Protocols for Quantum Entanglement Generation

In addition to an ever-growing list of applications in areas such as cybersecurity, medicine, and science, Machine Learning (ML) algorithms are also increasingly being applied to the field of quantum science, such as, e.g., quantum algorithms, quantum material science, quantum chemistry, quantum optics, and quantum many-body systems. We here investigate the potential of ML algorithms to drive progress in quantum information science, specifically quantum networks. In particular, we study if it is possible for an ML algorithm to self-learn optimal protocols for entanglement generation and distribution. Long-distance entanglement is a key requirement for quantum communication, specifically the realization of a long-distance quantum network (quantum internet). We will discuss the potential of using a projective-simulation-based reinforcement algorithm to identify successful entanglement generation protocols in noisy conditions.