Learning Equilibrium Strategies in Election Networks

Abstract

Social media has become in recent years an increasingly important domain for politics, especially for campaign advertising. In this work, we develop a tractable DeGroot model of influence maximization in political networks following initial advertisement in a two- candidate election from the viewpoint of a fully-informed social network platform. Our model considers two types of candidate objectives: margin of victory (maximizing total votes earned) and probability of victory (maximizing probability of earning the majority). We use it to show key theoretical differences in the corresponding games for arbitrarily large networks, including the existence of pure Nash equilibria. Finally, we contribute algorithms for computing mixed equilibria in the margin of victory case as well as influence- maximizing best-response algorithms in both cases, and implement them using the Adolescent Health Dataset.