AoI-Aware Resource Allocation for Platoon-Based C-V2X Networks via Multi-Agent Multi-Task Reinforcement Learning

The figure depicts two algorithms.Modified MADDPG which trains two critics. The global critic which is trained at the RSU and the local critic exclusive for each agent.Modified MADDPG with Task decomposition: This algorithm is similar to the Modified MADDPG; however, in this algorithm, the local holistic reward function of each agent is further decomposed into multiple sub-reward functions based on the tasks each agent has to accomplish, and the task-wise value functions are learned separately.