Prosocial Norm Emergence in Multi-agent Systems

Multi-agent systems provide a basis for developing of autonomous entities and thus find application in variety domains. We consider setting where not only the member agents are adaptive but also multi-agent system viewed as an entity its own right is adaptive. Specifically, social structure can be reflected norms among members. It well recognized that arise society always beneficial to focus on prosocial norms, which help achieve positive outcomes often guidance act manner takes into account welfare others. we propose Cha, framework emergence norms. Unlike previous norm approaches, Cha supports continual change (agents may enter leave) dynamism (norms when environment changes). Importantly, incorporate decision-making based inequity aversion theory, reflecting intuition guilt arising from being antisocial. In this manner, brings together two important themes prosociality: by individuals fairness system-level outcomes. demonstrate via simulation improve aggregate societal gains