Fairness through awareness

We study fairness in classification, where individuals are classified, e.g., admitted to a university, and the goal is to prevent discrimination against individuals based on their membership in some group, while maintaining utility for the classifier (the university). The main conceptual contribution of this paper is a framework for fair classification comprising (1) a (hypothetical) task-specific metric for determining the degree to which individuals are similar with respect to the classification task at hand; (2) an algorithm for maximizing utility subject to the fairness constraint, that similar individuals are treated similarly. We also present an adaptation of our approach to achieve the complementary goal of "fair affirmative action," which guarantees statistical parity (i.e., the demographics of the set of individuals receiving any classification are the same as the demographics of the underlying population), while treating similar individuals as similarly as possible. Finally, we discuss the relationship of fairness to privacy: when fairness implies privacy, and how tools developed in the context of differential privacy may be applied to fairness.

[1]  N. S. Barnett,et al.  Private communication , 1969 .

[2]  Harriett Tee Taggart,et al.  Redlining , 1981 .

[3]  M. Rabin Published by: American , 2022 .

[4]  Yuval Rabani,et al.  Fairness in scheduling , 1995, SODA '95.

[5]  Alexander V. Karzanov,et al.  Minimum 0-Extensions of Graph Metrics , 1998, Eur. J. Comb..

[6]  J. Rawls,et al.  Justice as Fairness: A Restatement , 2001 .

[7]  J. Buchanan John Rawls, Justice as Fairness: A Restatement , 2002 .

[8]  Éva Tardos,et al.  Approximation algorithms for classification problems with pairwise relationships: metric labeling and Markov random fields , 2002, JACM.

[9]  Joseph Naor,et al.  A Linear Programming Formulation and Approximation Algorithms for the Metric Labeling Problem , 2005, SIAM J. Discret. Math..

[10]  Jon M. Kleinberg,et al.  Metric embeddings with relaxed guarantees , 2005, 46th Annual IEEE Symposium on Foundations of Computer Science (FOCS'05).

[11]  Caterina Calsamiglia Decentralizing equality of opportunity and issues concerning the equality of educational opportunity , 2005 .

[12]  Cynthia Dwork,et al.  Differential Privacy , 2006, ICALP.

[13]  Nikhil Bansal,et al.  The Santa Claus problem , 2006, STOC '06.

[14]  Cynthia Dwork,et al.  Calibrating Noise to Sensitivity in Private Data Analysis , 2006, TCC.

[15]  Kunal Talwar,et al.  Mechanism Design via Differential Privacy , 2007, 48th Annual IEEE Symposium on Foundations of Computer Science (FOCS'07).

[16]  Uriel Feige,et al.  On allocations that maximize fairness , 2008, SODA '08.

[17]  Anupam Gupta,et al.  Approximating TSP on metrics with bounded global growth , 2008, SODA '08.

[18]  Behram F. T. Mistree,et al.  Gaydar: Facebook Friendships Expose Sexual Orientation , 2009, First Monday.

[19]  Kunal Talwar,et al.  On the geometry of differential privacy , 2009, STOC '10.

[20]  Moshe Tennenholtz,et al.  Mechanism design with uncertain inputs: (to err is human, to forgive divine) , 2011, STOC '11.