This paper focuses on distributed fault recovery in agent-based systems by providing help for faulty members. In the presented method, if one faulty agent requests for help or agents are informed of fault in one of th...
详细信息
This paper focuses on distributed fault recovery in agent-based systems by providing help for faulty members. In the presented method, if one faulty agent requests for help or agents are informed of fault in one of their teammates, they first decide if they are able to help or not. In the case that they are able to help and several help requests exist, helper agents specify a sequence of help actions through another distributed decision-making phase. The introduced fault clearing method is totally distributed in the sense that each helper agent makes its decisions by itself and no central or special agent exists in the system. In fact, the decision making process and the required information are designed such that the agents cooperate implicitly to prevent the system performance loss. The developed ideas are implemented in a simulated distributed control system. As it is shown, the proposed distributed fault-clearing method through reconfiguring the agents' roles is very effective.
In multiagent reinforcement learning, inter-agent credit assignment is a fundamental problem, since a single scalar reinforcement signal is the only reliable feedback that teams of learning agents receive. This proble...
详细信息
In multiagent reinforcement learning, inter-agent credit assignment is a fundamental problem, since a single scalar reinforcement signal is the only reliable feedback that teams of learning agents receive. This problem is more critical in groups of independent learners with a joint task. In this research, it is assumed that a critic agent receives the environment feedback and assigns a proper credit to each agent using some measures. Three of such measures for a team of cooperative agents with a parallel and AND-type task are introduced. These measures somehow compare the agents' knowledge. One of these criteria, called normal expertness, is a non-relative measure while two other ones (certainty and relative normal expertness) are relative measure. It is experimentally shown that relative measures work better as they contain more information for the critic agent.
暂无评论