Learning to Improve Coordinated Actions in Cooperative Distributed Problem-solving Environments 2. an Example Problem

Coordination is an essential technique in cooperative, distributed multi-agent systems. However, sophisticated coordination strategies are not always cost-effective in all problem-solving situations. This paper presents a learning method to identify what information will improve coordination in specific problem-solving situations. Learning is accomplished by recording and analyzing traces of inferences after problem solving. The analysis identifies situations in which inappropriate coordination strategies caused redundant activities or the lack of timely execution of important activities that degraded system performance. To remedy this problem, situation-specific control rules are created that acquire additional non-local information about activities in the agent networks and then select a presumably better plan or scheduling strategy. Examples from a real distributed problem-solving application involving diagnosis of a local area network are described. The content of the information does not necessarily reflect the position or the policy of the organizations supporting this research, and no official endorsement should be inferred. Achieving globally coherent 2 activity in a cooperative, distributed multi-agent system is a difficult problem for a number of reasons. One difficulty is that an agent's control decisions, based only on its local view of problem-solving task structures, may lead to inappropriate decisions about which activity it should do next, what results it should transmit to other agents or what results it should ask other agents to produce. If an agent has a view of the activities (task structures) of other agents, it can make more informed choices [4,6,7,13]. Another difficulty is that even with this type of meta-level information, there is still residual uncertainty about the outcomes of tasks and what future tasks will be coming into the system that may result in agents still exhibiting non-coherent behavior. These difficulties with achieving effective coordination are further exacerbated by the fact that an agent may expend significant computational resources in acquiring and exploiting a non-local view of other agents' activities. This expense is in terms of communication delays, as well as the computational cost of both providing this information in a suitable form to other agents and processing this information to make local decisions. Thus, for specific problem-solving situations, due to the inherent uncertainty in agents' activities and the cost of meta-level processing, it may not be worthwhile to acquire a complete view of other agents' activities, and thus some level of non-coherent activity may be necessary in order to achieve an optimal coordination strategy [13]. For example, coordination to …