Does mitigating ML's impact disparity require treatment disparity?

Abstract

Following related work in law and policy, two notions of disparity have come to shape the study of fairness in algorithmic decision-making. Algorithms exhibit treatment disparity if they formally treat members of protected subgroups differently; algorithms exhibit impact disparity when outcomes differ across subgroups, even if the correlation arises unintentionally. Naturally, we can achieve impact parity through purposeful treatment disparity. In one thread of technical work, papers aim to reconcile the two forms of parity proposing disparate learning processes (DLPs). Here, the learning algorithm can see group membership during training but produce a classifier that is group-blind at test time. In this paper, we show theoretically that: (i) When other features correlate to group membership, DLPs will (indirectly) implement treatment disparity, undermining the policy desiderata they are designed to address; (ii) When group membership is partly revealed by other features, DLPs induce within-class discrimination; and (iii) In general, DLPs provide a suboptimal trade-off between accuracy and impact parity. Based on our technical analysis, we argue that transparent treatment disparity is preferable to occluded methods for achieving impact parity. Experimental results on several real-world datasets highlight the practical consequences of applying DLPs vs. per-group thresholds.

Cited by in corpus (28)

• Fairness in Machine Learning: A Survey
• An Empirical Characterization of Fair Machine Learning For Clinical Risk Prediction
• Machine learning fairness notions: Bridging the gap with real-world applications
• Why Fairness Cannot Be Automated: Bridging the Gap Between EU Non-Discrimination Law and AI
• Can You Trust This Prediction? Auditing Pointwise Reliability After Learning
• It's COMPASlicated: The Messy Relationship between RAI Datasets and Algorithmic Fairness Benchmarks
• Predictive Multiplicity in Classification
• Noise-tolerant fair classification
• Explainability for fair machine learning
• A Distributionally Robust Approach to Fair Classification
• On Fair Selection in the Presence of Implicit and Differential Variance
• The Price of Local Fairness in Multistage Selection
• What does it mean to solve the problem of discrimination in hiring? Social, technical and legal perspectives from the UK on automated hiring systems
• Removing biased data to improve fairness and accuracy
• Fairness, Welfare, and Equity in Personalized Pricing
• Fair Machine Learning Under Partial Compliance
• Addressing multiple metrics of group fairness in data-driven decision making
• Neither Private Nor Fair: Impact of Data Imbalance on Utility and Fairness in Differential Privacy
• Explaining The Efficacy of Counterfactually Augmented Data
• To Split or Not to Split: The Impact of Disparate Treatment in Classification
• Toward a better trade-off between performance and fairness with kernel-based distribution matching
• Avoiding Resentment Via Monotonic Fairness
• Can Querying for Bias Leak Protected Attributes? Achieving Privacy With Smooth Sensitivity
• Fairness as Equality of Opportunity: Normative Guidance from Political Philosophy
• An example of prediction which complies with Demographic Parity and equalizes group-wise risks in the context of regression
• Towards Auditability for Fairness in Deep Learning
• Recommendation or Discrimination?: Quantifying Distribution Parity in Information Retrieval Systems
• On the impossibility of non-trivial accuracy under fairness constraints