Coalition formation is an important mechanism for cooperation in multiagent systems. In this paper we address the problem of coalition formation among self‐interested agents in superadditive task‐oriented domains. We assume that each agent has some “structure,” i.e., that it can be described by the values taken by a set ofmnonnegative attributes that represent the resourcesweach agent is endowed with. By defining the coalitional value as a functionVofw, we prove a sufficient condition for the existence of a stable payment configuration—in the sense of the core—in terms of certain properties ofV. We apply these ideas to a simple case that can be described by a linear program and show that it is possible to compute for it—in polynomial time—an optimal task allocation and a stable payment configuration.

Abstract In the last few years coalition formation algorithms have been proposed as a possible way of modeling autonomous agent cooperation in multi‐agent systems. This work is based on a previously proposed coalition formation model founded on game theory for a class of task‐oriented problems that guarantees an optimum task allocation and a stable profit division. In this paper we study two properties of the model that are very important for application in real‐life scenarios: robustness and tolerance to an agent's misbehavior. First, we study the robustness of this model as regards the effect the agent's failure has on the resultant profits of the coalition formation. Secondly, we also study the coalition formation model in the presence of misbehaving agents. Agents have some kind of execution autonomy, and they can deceive or mislead each other when they reveal their information, if they believe this will give them more profits. Copyright © 2005 John Wiley & Sons, Ltd.

In order to discover behavior patterns, current algorithms only analyze historical data in terms of performance data or fault events, ignoring the temporal correlation among different types of information, including the configuration changes. A method is presented that can discover recurrent patterns from multiple flows of events, such as alarms and configuration events, as well as discrete information, such as traffic and usage, taking into account static and dynamic information concerning observed objects and their environments. This method can filter out theoretically useless patterns, using a novel technique for detecting chaos in sequences of events. The prediction accuracy of the discovered patterns has been measured using objects with dynamic behavior controlled by known and complex differential equations. The proposed mining method has been used for discovering and predicting alarms in a computer network composed of several Internet servers taking into account the alarm and configuration events history, as well as static information about these servers.