Statistical Collusion by Collectives on Learning Platforms

Etienne Gauthier, Francis Bach, Michael I. Jordan

TL;DR: We study how collectives can pool their data to strategically modify it and influence learning platforms.

Abstract: As platforms increasingly rely on learning algorithms, collectives may form and seek ways to influence these platforms to align with their own interests. This can be achieved by coordinated submission of altered data. To evaluate the potential impact of such behavior, it is essential to understand the computations that collectives must perform to impact platforms in this way. In particular, collectives need to make a priori assessments of the effect of the collective before taking action, as they may face potential risks when modifying their data. Moreover they need to develop implementable coordination algorithms based on quantities that can be inferred from observed data. We develop a framework that provides a theoretical and algorithmic treatment of these issues and present experimental results in a product evaluation domain.

Lay Summary: Online platforms often use automated systems to make decisions based on user data. In response, groups of people might team up to try to influence these systems so that the outcomes better reflect their interests. They can do this by intentionally changing the information they provide to the platform. But before taking such steps, these groups need to figure out whether their efforts are likely to work. They also need practical ways to organize and act based on what they can observe. In this work, we build a framework that helps groups understand how to do this.

Link To Code: https://github.com/GauthierE/statistical-collusion

Primary Area: Social Aspects->Robustness

Keywords: Learning Algorithms, Collective Action, Data Poisoning

Submission Number: 6238