The Power of Context in Networks: Ideal Point Models with Social Interactions

Game theory has been widely used for modeling strategic behaviors in networked multiagent systems. However, the context within which these strategic behaviors take place has received limited attention. We present a model of strategic behavior in networks that incorporates the behavioral context. We focus on the contextual aspects of congressional voting. A senator's decision to vote yea or nay on a bill comes as a result of their ideologies, agendas, and their interactions with other senators. One salient model in political science is the ideal point model, which assigns each senator and each bill a number on the real line of political spectrum. These points then allow for prediction of future voting behavior. We extend the classical ideal point model with network-structured interactions among senators. In contrast to the ideal point model's prediction of individual voting behavior, we predict joint voting behaviors in a game-theoretic fashion. Our model also includes the characteristics of a bill. This allows it to outperform previous models that solely focus on the networked interactions among senators with no bill-specific parameters. We focus on two fundamental questions: learning the model using real-world data and computing stable outcomes of the model in order to predict joint voting behaviors. We demonstrate the effectiveness of our model through experiments using data from the 114th U.S. Congress.