Noyce Conference Room
Colloquium
US Mountain Time
Speaker:
Michael Kearns (University of Pennsylvania)
Our campus is closed to the public for this event.
Abstract. This talk will be a technical and self-contained counterpart to my SFI Community Lecture the evening before. I will begin by surveying recent frameworks for designing and analyzing “fair” machine learning algorithms and models, and then focus on fair learning via “chained” confidence intervals in sequential decision making, reinforcement learning, and cross-population ranking problems.