Is There a Trade-Off Between Fairness and Accuracy? A Perspective Using Mismatched Hypothesis Testing

Abstract

A trade-off between accuracy and fairness is almost taken as a given in the existing literature on fairness in machine learning. Yet, it is not preordained that accuracy should decrease with increased fairness. Novel to this work, we examine fair classification through the lens of mismatched hypothesis testing: trying to find a classifier that distinguishes between two ideal distributions when given two mismatched distributions that are biased. Using Chernoff information, a tool in information theory, we theoretically demonstrate that, contrary to popular belief, there always exist ideal distributions such that optimal fairness and accuracy (with respect to the ideal distributions) are achieved simultaneously: there is no trade-off. Moreover, the same classifier yields the lack of a trade-off with respect to ideal distributions while yielding a trade-off when accuracy is measured with respect to the given (possibly biased) dataset. To complement our main result, we formulate an optimization to find ideal distributions and derive fundamental limits to explain why a trade-off exists on the given biased dataset. We also derive conditions under which active data collection can alleviate the fairness-accuracy trade-off in the real world. Our results lead us to contend that it is problematic to measure accuracy with respect to data that reflects bias, and instead, we should be considering accuracy with respect to ideal, unbiased data.

arXiv commentThis paper appears in the Proceedings of the 37th International Conference on Machine Learning, pp. 2803--2813, 2020

Cited by in corpus (15)

• Oort: Efficient Federated Learning via Guided Participant Selection
• Inherent Tradeoffs in Learning Fair Representations
• FairFaceGAN: Fairness-aware Facial Image-to-Image Translation
• Revisiting Chernoff Information with Likelihood Ratio Exponential Families
• On Fair Selection in the Presence of Implicit and Differential Variance
• How to be fair? A study of label and selection bias
• Fundamental Limits and Tradeoffs in Invariant Representation Learning
• Causally motivated Shortcut Removal Using Auxiliary Labels
• Pareto Efficient Fairness in Supervised Learning: From Extraction to Tracing
• Can Information Flows Suggest Targets for Interventions in Neural Circuits?
• Where Is the Normative Proof? Assumptions and Contradictions in ML Fairness Research
• Removing Spurious Features can Hurt Accuracy and Affect Groups Disproportionately
• Fair Representation Learning using Interpolation Enabled Disentanglement
• A Sociotechnical View of Algorithmic Fairness
• Pairwise Fairness for Ordinal Regression