A Mean-Field Stackelberg Game Approach for Obfuscation Adoption in Empirical Risk Minimization
This addresses privacy concerns in large-scale data systems like online tracking and IoT, offering a strategic solution for balancing user obfuscation and algorithm accuracy, though it is incremental in applying existing game theory concepts to this domain.
The paper tackles the problem of privacy conflicts in data ecosystems by developing a game-theoretic framework that combines Stackelberg and mean field games to model interactions between users and tracking algorithms, resulting in equilibrium conditions that incentivize obfuscation and privacy protection.
Data ecosystems are becoming larger and more complex due to online tracking, wearable computing, and the Internet of Things. But privacy concerns are threatening to erode the potential benefits of these systems. Recently, users have developed obfuscation techniques that issue fake search engine queries, undermine location tracking algorithms, or evade government surveillance. Interestingly, these techniques raise two conflicts: one between each user and the machine learning algorithms which track the users, and one between the users themselves. In this paper, we use game theory to capture the first conflict with a Stackelberg game and the second conflict with a mean field game. We combine both into a dynamic and strategic bi-level framework which quantifies accuracy using empirical risk minimization and privacy using differential privacy. In equilibrium, we identify necessary and sufficient conditions under which 1) each user is incentivized to obfuscate if other users are obfuscating, 2) the tracking algorithm can avoid this by promising a level of privacy protection, and 3) this promise is incentive-compatible for the tracking algorithm.