A Mean-Field Stackelberg Game Approach for Obfuscation Adoption in Empirical Risk Minimization

TL;DR

This paper models the strategic interactions between users and tracking algorithms in data ecosystems using a combined mean-field and Stackelberg game framework, analyzing obfuscation incentives and privacy-accuracy trade-offs.

Contribution

It introduces a novel bi-level game-theoretic framework integrating mean-field and Stackelberg games to analyze obfuscation and privacy in empirical risk minimization.

Findings

Users are incentivized to obfuscate when others do so.

Tracking algorithms can prevent obfuscation by offering privacy guarantees.

Optimal privacy promises are incentive-compatible for algorithms.

Abstract

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.