Representation and Learning in Computational Game Theory

Game theory has emerged as the key tool for understanding and designing complex multiagent environments such as the Internet, systems of autonomous agents, and electronic communities or economies. To support these relatively recent uses of game theory — as well as for more ambitious modeling in some of the traditional application areas — there is a growing need for a computational theory that is largely absent from classical game theory research. Such a computational theory needs to provide a rich and flexible collection of models and representations for complex game-theoretic problems; powerful and efficient algorithms for manipulating and learning these models; and a deep understanding of the algorithmic and resource issues arising in all aspects of game theory. The overarching goal of the proposed work is to “scale up” the applicability of game theory, in much the same way that Bayesian networks and associated advances made complex, highdimensional probabilistic modeling possible in a wide set of applications in computer science and beyond. Two of the most important topics that have materialized to date — and the primary emphases of the current proposal — are the representation and efficient manipulation of large and complex games, and new approaches to learning in game-theoretic settings. On the topic of representation, the proposal includes the development of methods to model structured interaction in large-population games; the intersection of social network theory and game theory; new representations in repeated games; and representational issues for a variety of equilibria types. On the topic of learning, it includes the development of online multiplicative update methods for large and structured games; modeling cooperation in learning; applications of game theory to the analysis of machine learning methods; learning for games that change over time; and the relationship between game theory and reinforcement learning. It also covers many interesting topics in the intersection of representational and learning issues. The work proposed here will have a number of broader impacts. It will make it possible to study interactions in social and ecological systems in drastically larger systems using the new game-theoretic representations and algorithms, allowing researchers in other scientific disciplines to apply game theory to a broader array of problems. Through highly visible symposia, workshops, and tutorials at international conferences and interdisciplinary classes on computational game theory, the new discoveries will be disseminated to interested researchers and students. In addition, new programs in game theory will be created for two established outreach programs based at Penn’s Institute for Research in Cognitive Science. One of these is a summer school for advanced undergraduates, while the other is for minority and female high school students in the Philadelphia area.