在多个智能体存在的场景，智能体们以合作的方式完成一个任务（最大化一个全

In a scenario where multiple agents exist, the agents complete a task in a cooperative manner (maximizing a global cumulative reward), and users who access the same channel can interact with each other. Each agent is required to consider The state of self-observation must also take into account the actions and strategies of other agents to learn a strategy that maximizes the global cumulative reward

In the scene where multiple agents exist, Agents complete a task in a cooperative way (maximizing a global cumulative reward). Users accessing the same channel can interact with each other. Each agent is required to consider not only its own observed state, but also the actions, strategies and other information of other agents, so as to learn a strategy to maximize the global cumulative reward

In the scene where multiple agents exist, agents complete a task in a cooperative way (maximizing a global cumulative reward), and users accessing the same channel can interact with each other. It is required that each agent not only consider the observed state of itself, but also consider the actions and strategies of other agents, so as to learn a strategy to maximize the global cumulative reward.