Algorithmic Fairness: Choices, Assumptions, and Definitions

2018/11/30 by Shira Mitchell, Eric Potash, Solon Barocas +3 · 3 citations
Computer Science · Mathematics · Social Sciences · #Adversarial Robustness in Machine Learning #Computer science #Data science #Economics #Ethics and Social Impacts of AI #Explainable Artificial Intelligence (XAI) #Field (mathematics) #Management science #Mathematics #Notation #Operations research #Order (exchange) #Terminology #stat.AP

paper · pdf · doi:10.1146/annurev-statistics-042720-125902

published as Annual Review of Statistics and Its Application 2021 8:1

arxiv created 2020/04/24 · openalex publication_date 2020/11/10 · arxiv updated 2020/11/23 · openalex created_date 2020/11/23 · openalex updated_date 2026/08/05

Abstract

A recent wave of research has attempted to define fairness quantitatively. In particular, this work has explored what fairness might mean in the context of decisions based on the predictions of statistical and machine learning models. The rapid growth of this new field has led to wildly inconsistent motivations, terminology, and notation, presenting a serious challenge for cataloging and comparing definitions. This article attempts to bring much-needed order. First, we explicate the various choices and assumptions made—often implicitly—to justify the use of prediction-based decision-making. Next, we show how such choices and assumptions can raise fairness concerns and we present a notationally consistent catalog of fairness definitions from the literature. In doing so, we offer a concise reference for thinking through the choices, assumptions, and fairness considerations of prediction-based decision-making.

Citations

• Statistical Decision Theory and Bayesian Analysis
• Certifying and Removing Disparate Impact
• The Theory of Decision Procedures for Distributions with Monotone Likelihood Ratio
• Fairness through awareness
• Equality of Opportunity in Supervised Learning
• RISK, RACE, AND RECIDIVISM: PREDICTIVE BIAS AND DISPARATE IMPACT*
• Fairness in Criminal Justice Risk Assessments: The State of the Art
• Counterfactual Fairness
• Preventing Fairness Gerrymandering: Auditing and Learning for Subgroup Fairness
• Probably Approximately Metric-Fair Learning
• Fairness Through Computationally-Bounded Awareness
• Residual Unfairness in Fair Machine Learning from Prejudiced Data
• A Broader View on Bias in Automated Decision-Making: Reflecting on Epistemology and Dynamics
• Welfare and Distributional Impacts of Fair Classification
• Model Cards for Model Reporting
• Fairness and Abstraction in Sociotechnical Systems
• The Frontiers of Fairness in Machine Learning
• Failing Loudly: An Empirical Study of Methods for Detecting Dataset Shift
• 50 Years of Test (Un)fairness: Lessons for Machine Learning
• An algorithm for removing sensitive information: application to race-independent recidivism prediction
• Inherent Trade-Offs in the Fair Determination of Risk Scores

Cited by

• Explainability and justification of automatic-decision making: A conceptual framework and a practical application
• Local Statistical Parity for the Estimation of Fair Decision Trees
• A Systematic Approach to Group Fairness in Automated Decision Making