Tolerance is Necessary for Stability: Single-Peaked Swap Schelling Games

Residential segregation in metropolitan areas is a phenomenon that can be observed all over the world. It is characterized by the emergence of large regions populated by residents that are homogeneous in terms of ethnicity or other traits. In a recent research trend in the AI community this phenomenon was investigated via game-theoretic models. There, selfish agents of two types are equipped with a monotone utility function that ensures higher utility if an agent has more same-type neighbors. The agents strategically choose their location on a given graph that serves as residential area to maximize their utility. However, sociological polls suggest that real-world agents are actually favoring mixed-type neighborhoods, and hence should be modeled via non-monotone utility functions. We study Swap Schelling Games with non-monotone utility functions that are single-peaked. In these games pairs of agents may improve their utility by swapping their locations. Our main finding is that tolerance, i.e., that the agents favor fifty-fifty neighborhoods or even being in the minority, is necessary for equilibrium existence on almost regular or bipartite graphs. We show equilibrium existence on almost regular graphs via a potential function argument and we prove that this approach is impossible on arbitrary graphs even with tolerant agents. Regarding the quality of equilibria, we consider the recently introduced degree of integration, that counts the number of agents that live in a heterogeneous neighborhood, as social welfare function. We derive (almost) tight bounds on the Price of Anarchy and the Price of Stability. In particular, we show that the latter is constant on bipartite and almost regular graphs. Moreover, we prove that computing approximations of the social optimum placement and the equilibrium with maximum social welfare is NP-hard even on cubic graphs.