A Reductions Approach to Fair Classification

Abstract

We present a systematic approach for achieving fairness in a binary classification setting. While we focus on two well-known quantitative definitions of fairness, our approach encompasses many other previously studied definitions as special cases. The key idea is to reduce fair classification to a sequence of cost-sensitive classification problems, whose solutions yield a randomized classifier with the lowest (empirical) error subject to the desired constraints. We introduce two reductions that work for any representation of the cost-sensitive classifier and compare favorably to prior baselines on a variety of data sets, while overcoming several of their disadvantages.

References in corpus (3)

• Equality of Opportunity in Supervised Learning
• Preventing Fairness Gerrymandering: Auditing and Learning for Subgroup Fairness
• Empirical Risk Minimization under Fairness Constraints

Cited by in corpus (50)

• Fairness in Machine Learning: A Survey
• User-oriented Fairness in Recommendation
• Personalized Counterfactual Fairness in Recommendation
• Connecting User and Item Perspectives in Popularity Debiasing for Collaborative Recommendation
• Identifying and Correcting Label Bias in Machine Learning
• Empirical Risk Minimization under Fairness Constraints
• Bias in Machine Learning Models Can Be Significantly Mitigated by Careful Training: Evidence from Neuroimaging Studies
• Value Cards: An Educational Toolkit for Teaching Social Impacts of Machine Learning through Deliberation
• Medical Imaging and Machine Learning
• On Adversarial Bias and the Robustness of Fair Machine Learning
• Factors Influencing Perceived Fairness in Algorithmic Decision-Making: Algorithm Outcomes, Development Procedures, and Individual Differences
• Learning Certified Individually Fair Representations
• Uncertainty as a Form of Transparency: Measuring, Communicating, and Using Uncertainty
• Improved Adversarial Learning for Fair Classification
• To be Robust or to be Fair: Towards Fairness in Adversarial Training
• Review of Mathematical frameworks for Fairness in Machine Learning
• Achieving Fairness in the Stochastic Multi-armed Bandit Problem
• Towards Fair Knowledge Transfer for Imbalanced Domain Adaptation
• Differentially Private Fair Learning
• FAT Forensics: A Python Toolbox for Algorithmic Fairness, Accountability and Transparency
• ESR: Ethics and Society Review of Artificial Intelligence Research
• Rényi Fair Inference
• Fairness On The Ground: Applying Algorithmic Fairness Approaches to Production Systems
• TFROM: A Two-sided Fairness-Aware Recommendation Model for Both Customers and Providers
• Bias, Fairness, and Accountability with AI and ML Algorithms
• Multi-objective Asynchronous Successive Halving
• Addressing multiple metrics of group fairness in data-driven decision making
• Does enforcing fairness mitigate biases caused by subpopulation shift?
• Black Loans Matter: Distributionally Robust Fairness for Fighting Subgroup Discrimination
• On the Privacy Risks of Algorithmic Fairness
• Unbiased Subdata Selection for Fair Classification: A Unified Framework and Scalable Algorithms
• Keeping Designers in the Loop: Communicating Inherent Algorithmic Trade-offs Across Multiple Objectives
• Maximum Weighted Loss Discrepancy
• Wasserstein Robust Classification with Fairness Constraints
• Fairness through Experimentation: Inequality in A/B testing as an approach to responsible design
• Fair Classification and Social Welfare
• Multi-Differential Fairness Auditor for Black Box Classifiers
• Learning Smooth and Fair Representations
• User Acceptance of Gender Stereotypes in Automated Career Recommendations
• Towards a Measure of Individual Fairness for Deep Learning
• Measuring Fairness Under Unawareness of Sensitive Attributes: A Quantification-Based Approach
• Identifying biases in legal data: An algorithmic fairness perspective
• Hidden Technical Debts for Fair Machine Learning in Financial Services
• OmniFair: A Declarative System for Model-Agnostic Group Fairness in Machine Learning
• Bias-Tolerant Fair Classification
• Fairness constraints can help exact inference in structured prediction
• Towards Auditability for Fairness in Deep Learning
• Adversarial Robustness Across Representation Spaces
• Group-based Fair Learning Leads to Counter-intuitive Predictions
• Practical Compositional Fairness: Understanding Fairness in Multi-Component Recommender Systems