Quantum Reinforcement Learning: Concepts, Models, and Applications

This tutorial presents the fundamental theory of quantum reinforcement learning (QRL) and its emerging applications. Thanks to the recent evolution in quantum computing, numerous research results have been proposed for the development of theories and applications of QRL. In this tutorial paper, the advantages and benefits of QRL-based models and algorithms are also discussed, i.e., fast learning convergence, high action-dimension scalability, and efficient training parameter utilization. Based on the design concept and advantages, QRL-based models and algorithms are used for various emerging applications. Among them, this paper introduces the applications of QRL-based models and algorithms in terms of future multimedia systems, autonomous mobility services, and distributed computing platforms.