BUSEC Distinguished Lecture: Cynthia Dwork

Starts:
11:00 am on Wednesday, May 1, 2013
“Why was I not shown this advertisement? Why was my loan application denied? Why was I rejected from this university?” This talk will address fairness in classification, where the goal is to prevent discrimination against protected population subgroups in classification systems while simultaneously preserving utility for the party carrying out the classification, for example, the advertiser, bank, or admissions committee. We argue that a classification is fair only when individuals who are similar with respect to the classification task at hand are treated similarly, and this in turn requires understanding of sub cultures of the population. Similarity metrics are applied in many contexts, but these are often hidden. Our work explicitly exposes the metric, opening it to public debate. (Joint work with Moritz Hardt, Toniann Pitassi, Omer Reingold, and Richard Zemel.) Our approach provides a (theoretical) method by which an on-line advertising network can prevent discrimination against protected groups, even when the advertisers are unknown and untrusted. We briefly discuss the role of fairness in consumer objections to behavioral targeting and explain how traditional notions of privacy miss the mark and fail to address these. (Joint work with Deirdre Mulligan.) Finally, we discuss a machine learning instantiation of our approach, in which the distance metric need not be given but can instead be learned. (Joint work with Toniann Pitassi, Yu Wu, and Richard Zemel.)