Auditing ML Models for Individual Bias and Unfairness

2020/03/11 by Songkai Xue, Mikhail Yurochkin, Xue, Songkai +3 · 4 citations
Computer Science · Social Sciences · #Adversarial Robustness in Machine Learning #Ethics and Social Impacts of AI #Explainable Artificial Intelligence (XAI) #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML)

paper · pdf · doi:10.48550/arxiv.2003.05048

openalex publication_date 2020/03/11 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

We consider the task of auditing ML models for individual bias/unfairness. We formalize the task in an optimization problem and develop a suite of inferential tools for the optimal value. Our tools permit us to obtain asymptotic confidence intervals and hypothesis tests that cover the target/control the Type I error rate exactly. To demonstrate the utility of our tools, we use them to reveal the gender and racial biases in Northpointe's COMPAS recidivism prediction instrument.

Citations

• Bootstrap Methods: Another Look at the Jackknife
• Certifying Some Distributional Robustness with Principled Adversarial\n Training
• Minimax Statistical Learning with Wasserstein Distances
• Confidence Regions in Wasserstein Distributionally Robust Estimation

Cited by

• Wasserstein projection distance for fairness testing of regression models
• Cost Efficient Fairness Audit Under Partial Feedback
• From Fragile to Certified: Wasserstein Audits of Group Fairness Under Distribution Shift
• Everything is Relative: Understanding Fairness with Optimal Transport

Related

• Statistical inference for individual fairness
• Statistical Inference for Fairness Auditing
• DispaRisk: Auditing Fairness Through Usable Information
• Visual Auditor: Interactive Visualization for Detection and Summarization of Model Biases
• Detecting Statistically Significant Fairness Violations in Recidivism Forecasting Algorithms