Reinforcement Learning and Deep Reinforcement Learning for AoI-Aware UAV-IoT
摘要
This chapter explores reinforcement learning (RL) and deep reinforcement learning (DRL) algorithms, both in general and in the context of Age of Information (AoI)-aware UAV-assisted WSN/IoT networks. It discusses the three main classes of RL algorithms: value-based, policy-based, and actor-critic. A range of DRL solutions and algorithms designed to learn optimal policies for some of the Markov Decision Processes (MDPs) introduced in Chap. 2 are examined in this chapter. These DRL approaches are implemented in both single-agent and multi-agent environments.