Fairness Through Awareness

Abstract

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.

Cited by in corpus (22)

• Aligning AI With Shared Human Values
• Causal Mediation Analysis for Interpreting Neural NLP: The Case of Gender Bias
• FairGAN: Fairness-aware Generative Adversarial Networks
• The Fairness of Risk Scores Beyond Classification: Bipartite Ranking and the xAUC Metric
• Noise-tolerant fair classification
• AI Research Considerations for Human Existential Safety (ARCHES)
• Two Simple Ways to Learn Individual Fairness Metrics from Data
• Exactly Computing the Local Lipschitz Constant of ReLU Networks
• SenSeI: Sensitive Set Invariance for Enforcing Individual Fairness
• FairVis: Visual Analytics for Discovering Intersectional Bias in Machine Learning
• ShapeFlow: Dynamic Shape Interpreter for TensorFlow
• Digital Voodoo Dolls
• Does enforcing fairness mitigate biases caused by subpopulation shift?
• Black Loans Matter: Distributionally Robust Fairness for Fighting Subgroup Discrimination
• An Information-theoretic Approach to Distribution Shifts
• A Decentralized Approach towards Responsible AI in Social Ecosystems
• Getting Fairness Right: Towards a Toolbox for Practitioners
• Equalizing Financial Impact in Supervised Learning
• Towards the Right Kind of Fairness in AI
• Fairness-Aware Online Meta-learning
• Heuristic-Based Weak Learning for Automated Decision-Making
• Reducing Discrimination in Learning Algorithms for Social Good in Sociotechnical Systems