As algorithms reach ever more deeply into our daily lives, determining our online experiences, credit offerings, and educational opportunities, and contributing to decisions on eligibility for bail and parole, there is increasing concern that they be “fair”. But what does “fair” mean? How can fairness be ensured? When can we conclude that a system composed of pieces that are fair in isolation is fair in toto? The exploration of these questions is giving rise to the emerging theory of algorithmic fairness. This cluster brings together architects of this new area and researchers in machine learning of fair representations.


Long-Term Participants (including Organizers)

Visiting Graduate Students and Postdocs