Fair algorithms via game theory and sequential learning – FairPlay

Machine learning algorithms are increasingly used to optimize decision making in various areas, but this can result in unacceptable discrimination. The main objective of this project is to propose an innovative framework for the development of learning algorithms that respect fairness constraints. While the literature mostly focuses on idealized settings, the originality of this framework and central focus of this project is the use of game theory and sequential learning methods to account for constraints that appear in practical applications: strategic and decentralized aspects of the decisions and the data provided and absence of knowledge of certain parameters key to the fairness definition. This framework will allow the development of learning algorithms that are optimal in a regret-theoretic sense while respecting fairness constraints. To demonstrate the effectiveness of our approach, we propose to apply this framework in the contexts of auction-based systems (e.g., online targeted advertising) and selection problems (e.g., hiring or college admission).