Algorithmic Game Theory and Applications

Methods from game theory and mechanism design have been proven to be a powerful mathematical tool in order to understand, control, and efficiently design dynamic, complex networks, such as the Internet. Game theory provides a good starting point for computer scientists to understand selfish rational behavior of complex networks with many agents. Such a scenario is readily modeled using game theory techniques, in which players with potentially different goals participate under a common setting with well prescribed interactions. The Nash equilibrium stands out as the predominant concept of rationality in noncooperative settings. Thus, game theory and its notions of equilibria provide a rich framework for modeling the behavior of selfish agents in these kinds of distributed and networked environments and offering mechanisms to achieve efficient and desirable global outcomes despite selfish behavior. The most important algorithmic solutions and advances achieved through game theory are reviewed.