Abstract
Much research in MAS explores how refinements to one agent?s reasoning can improve system performance. Sometimes, aspects of a system?s behavior are independent of individual agents? algorithms. Inspired by statistical physics, we term this phenomenon "universality": systems whose elements differ widely may have common emergent features. We develop a notion of universality in MAS based on the concept?s use in its original (physics) setting. We give examples of the phenomenon, and discuss its implications for the theory and practice of MAS. We speculate that there exists a hierarchy of types of universality. The statistical mechanics sense refers to the most refined, simplest, and quantitative, while commonalities among MAS systems are associated with somewhat more general and qualitative levels of universality. Such a hierarchy would be an important integrating principle across systems of interacting components, including human societies, animal ecologies, multi-agent systems, and atoms and molecules.