Simplicity Creates Inequity: Implications for Fairness, Stereotypes, and InterpretabilityEI
Algorithms can be a powerful aid to decision-making - particularly when decisions rely, even implicitly, on predictions . We are already seeing algorithms play this role in domains including hiring, education, lending, medicine, and criminal justice [2, 6, 10]. As is typical in machine learning applications, accuracy is an important measure for these tasks.
- 1Paul R Milgrom, Good Nevs and Bad News: Representation Theorems and Applications., 1979.
- 2Anthony G Greenwald, Mahzarin R Banaji, Anthony G Grtenwald, Washington Mahzarin, R Banaji. Implicit social cognition: Attitudes, self-esteem, and stereotypes.Psychological Review, pp. 4-27, 1995.
- 5Susan Athey, Monotone Comparative Statics Under Uncertainty.The American Economic Review, 2002.
- 6Hanming Fang, Andrea Moro. Theories of Statistical Discrimination and Affirmative Action: A Survey., 2010.
- 10Bryce Goodman, Seth R. Flaxman. European Union Regulations on Algorithmic Decision-Making and a "Right to Explanation".AI Magazine, pp. 50-57, 2017.
- 13Sam Corbett-Davies, Emma Pierson, Avi Feller, Sharad Goel, Aziz Huq. Algorithmic decision making and the cost of fairness.Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2017.
- 14Finale Doshi-Velez, Been Kim. A Roadmap for a Rigorous Science of Interpretability.CoRR, 2017.
- 16Megan T. Stevenson, Assessing Risk Assessment in Action.Social Science Research Network, 2017.
- 18Zachary C. Lipton, Julian McAuley, Alexandra Chouldechova. Does mitigating ML's impact disparity require treatment disparity?.NeurIPS, pp. 8136-8146, 2018.
- 19Sam Corbett-Davies, Sharad Goel. The Measure and Mismeasure of Fairness: A Critical Review of Fair Machine Learning.CoRR, 2018.