Prediction-Based Decisions and Fairness: A Catalogue of Choices, Assumptions, and Definitions

Abstract

A recent flurry of research activity has attempted to quantitatively define "fairness" for decisions based on statistical and machine learning (ML) predictions. The rapid growth of this new field has led to wildly inconsistent terminology and notation, presenting a serious challenge for cataloguing and comparing definitions. This paper attempts to bring much-needed order. First, we explicate the various choices and assumptions made---often implicitly---to justify the use of prediction-based decisions. Next, we show how such choices and assumptions can raise concerns about fairness and we present a notationally consistent catalogue of fairness definitions from the ML literature. In doing so, we offer a concise reference for thinking through the choices, assumptions, and fairness considerations of prediction-based decision systems.

References in corpus (27)

• Model Cards for Model Reporting
• Equality of Opportunity in Supervised Learning
• Counterfactual Fairness
• Fairness Beyond Disparate Treatment & Disparate Impact: Learning Classification without Disparate Mistreatment
• Using stacking to average Bayesian predictive distributions
• The Measure and Mismeasure of Fairness
• Inherent Trade-Offs in the Fair Determination of Risk Scores
• Fairness Testing: Testing Software for Discrimination
• Preventing Fairness Gerrymandering: Auditing and Learning for Subgroup Fairness
• Fairness Constraints: Mechanisms for Fair Classification
• 50 Years of Test (Un)fairness: Lessons for Machine Learning
• On the (im)possibility of fairness
• The Frontiers of Fairness in Machine Learning
• Problem Formulation and Fairness
• Failing Loudly: An Empirical Study of Methods for Detecting Dataset Shift
• Avoiding Discrimination through Causal Reasoning
• A Short-term Intervention for Long-term Fairness in the Labor Market
• A Nutritional Label for Rankings
• Fairness Through Computationally-Bounded Awareness
• Residual Unfairness in Fair Machine Learning from Prejudiced Data
• Probably Approximately Metric-Fair Learning
• A Complete Generalized Adjustment Criterion
• A Broader View on Bias in Automated Decision-Making: Reflecting on Epistemology and Dynamics
• Fairer and more accurate, but for whom?
• Welfare and Distributional Impacts of Fair Classification
• Causal Diagrams for Interference
• An algorithm for removing sensitive information: application to race-independent recidivism prediction

Cited by in corpus (45)

• The Fallacy of AI Functionality
• A Clarification of the Nuances in the Fairness Metrics Landscape
• Fairness in Credit Scoring: Assessment, Implementation and Profit Implications
• Rethinking Fairness: An Interdisciplinary Survey of Critiques of Hegemonic ML Fairness Approaches
• Algorithms and Decision-Making in the Public Sector
• Fairness and Bias in Algorithmic Hiring: a Multidisciplinary Survey
• Metrics for Dataset Demographic Bias: A Case Study on Facial Expression Recognition
• `It is currently hodgepodge'': Examining AI/ML Practitioners' Challenges during Co-production of Responsible AI Values
• On the Moral Justification of Statistical Parity
• Regression Diagnostics meets Forecast Evaluation: Conditional Calibration, Reliability Diagrams, and Coefficient of Determination
• Algorithmic Fairness and Structural Injustice: Insights from Feminist Political Philosophy
• The Algorithmic Imprint
• Measuring Disparate Outcomes of Content Recommendation Algorithms with Distributional Inequality Metrics
• Precision psychiatry: predicting predictability
• A Classification of Feedback Loops and Their Relation to Biases in Automated Decision-Making Systems
• Augmented Datasheets for Speech Datasets and Ethical Decision-Making
• Reinforcement Learning in Modern Biostatistics: Constructing Optimal Adaptive Interventions
• Can Fairness be Automated? Guidelines and Opportunities for Fairness-aware AutoML
• Bridging the gap: Towards an Expanded Toolkit for AI-driven Decision-Making in the Public Sector
• Integrating Psychometrics and Computing Perspectives on Bias and Fairness in Affective Computing: A Case Study of Automated Video Interviews
• Fairness in Agreement With European Values: An Interdisciplinary Perspective on AI Regulation
• Justice in Misinformation Detection Systems: An Analysis of Algorithms, Stakeholders, and Potential Harms
• Enforcing Group Fairness in Algorithmic Decision Making: Utility Maximization Under Sufficiency
• Multi-Target Multiplicity: Flexibility and Fairness in Target Specification under Resource Constraints
• Potential for allocative harm in an environmental justice data tool
• It's Not You, It's Me: The Impact of Choice Models and Ranking Strategies on Gender Imbalance in Music Recommendation
• Fairness Hacking: The Malicious Practice of Shrouding Unfairness in Algorithms
• SoK: Machine Learning Governance
• Group Fairness in Prediction-Based Decision Making: From Moral Assessment to Implementation
• On Prediction-Modelers and Decision-Makers: Why Fairness Requires More Than a Fair Prediction Model
• Materiality and Risk in the Age of Pervasive AI Sensors
• From the Fair Distribution of Predictions to the Fair Distribution of Social Goods: Evaluating the Impact of Fair Machine Learning on Long-Term Unemployment
• Access Denied: Meaningful Data Access for Quantitative Algorithm Audits
• Robustness and Cybersecurity in the EU Artificial Intelligence Act
• Forks Over Knives: Predictive Inconsistency in Criminal Justice Algorithmic Risk Assessment Tools
• Auditing Cross-Cultural Consistency of Human-Annotated Labels for Recommendation Systems
• Formalising Anti-Discrimination Law in Automated Decision Systems
• Gerrymandering Individual Fairness
• SCALES: From Fairness Principles to Constrained Decision-Making
• Multiversal Simulacra: Understanding Hypotheticals and Possible Worlds Through Simulation
• A Systematic Approach to Group Fairness in Automated Decision Making
• Individual Fairness for Social Media Influencers
• Talking About the Assumption in the Room
• Searching Personal Collections
• Causal Models in Requirement Specifications for Machine Learning: A vision