Quantifying Outcomes in Games

Network games play a fundamental role in understanding behavior in many domains, ranging from communication networks through markets to social networks. In this talk we'll study the degradation of quality of solution caused by the selfish behavior of users in a number of different games including congestion games that model routing or cost-sharing, and games that model Ad-Auctions. In each setting our goal is to quantify the degradation of quality of solution caused by the selfish behavior of user. We compare the selfish outcome to a centrally designed optimum both in terms of the quality of Nash equilibria and also the quality of outcomes of learning behavior by the users.