An Adaptive Control Technique for a Connection Weight of Agents in a Self-repairing Network

Abstract

Cooperation among agents is a crucial problem in autonomous distributed systems composed of selfish agents pursuing their own profits. An earlier study of a self-repairing network revealed that a systemic payoff made the selfish agents cooperate with other agents and was similar to kin selection. We study the relationship between the systemic payoff and kin selection more deeply. This paper considers the systemic payoff that involves a connection weight representing strength of relationship among the agents. We found that the performance of the self-repairing network changes by varying the connection weight. The connection weight appropriate to the environments would elicit the good performance of the self-repairing network. This paper proposes an adaptive control technique for the connection weight among the agents in the systemic payoff. The technique changes the connection weight dynamically. In simulations, the proposed technique showed the good performance in the dynamic environments.Keywordsautonomous distributed systemskin selectiongame theoryselfish agentsself-repairing network