Minimax Pareto Fairness: A Multi Objective Perspective

Abstract

In this work we formulate and formally characterize group fairness as a multi-objective optimization problem, where each sensitive group risk is a separate objective. We propose a fairness criterion where a classifier achieves minimax risk and is Pareto-efficient w.r.t. all groups, avoiding unnecessary harm, and can lead to the best zero-gap model if policy dictates so. We provide a simple optimization algorithm compatible with deep neural networks to satisfy these constraints. Since our method does not require test-time access to sensitive attributes, it can be applied to reduce worst-case classification errors between outcomes in unbalanced classification problems. We test the proposed methodology on real case-studies of predicting income, ICU patient mortality, skin lesions classification, and assessing credit risk, demonstrating how our framework compares favorably to other approaches.

Cited by in corpus (10)

• Out of Context: Investigating the Bias and Fairness Concerns of "Artificial Intelligence as a Service"
• Robustness Disparities in Commercial Face Detection
• Comparing Human and Machine Bias in Face Recognition
• Pareto Efficient Fairness in Supervised Learning: From Extraction to Tracing
• Technical Challenges for Training Fair Neural Networks
• How Robust is Your Fairness? Evaluating and Sustaining Fairness under Unseen Distribution Shifts
• A Hybrid 2-stage Neural Optimization for Pareto Front Extraction
• Federating for Learning Group Fair Models
• The Sharpe predictor for fairness in machine learning
• A Unified Approach to Fair Online Learning via Blackwell Approachability