Emergence of social norms through collective learning in networked agent societies
Social norms play a pivotal role in sustaining social order by regulating individual behaviors in a society. In normative multiagent systems, social norms have been used as an efficient mechanism to govern virtual agent societies towards cooperation and coordination. In this paper, we study the emergence of social norms via learning from repeated local interactions in networked agent societies. We propose a collective learning framework, which imitates the opinion aggregation process in human decision making, to study the impact of agent local collective behaviors on norm emergence in different situations. In the framework, each agent interacts repeatedly with all of its neighbors. At each step, an agent first takes a best-response action towards each of its neighbors and then combines all of these actions into a final action using ensemble learning methods. We conduct extensive experiments to evaluate the framework with respect to different network topologies, learning strategies, numbers of actions, and so on. Experimental results reveal some significant insights into norm emergence in networked agent societies achieved through local collective behaviors.