A Unified Approach to Quantifying Algorithmic Unfairness: Measuring Individual & Group Unfairness via Inequality Indices

Abstract

Discrimination via algorithmic decision making has received considerable attention. Prior work largely focuses on defining conditions for fairness, but does not define satisfactory measures of algorithmic unfairness. In this paper, we focus on the following question: Given two unfair algorithms, how should we determine which of the two is more unfair? Our core idea is to use existing inequality indices from economics to measure how unequally the outcomes of an algorithm benefit different individuals or groups in a population. Our work offers a justified and general framework to compare and contrast the (un)fairness of algorithmic predictors. This unifying approach enables us to quantify unfairness both at the individual and the group level. Further, our work reveals overlooked tradeoffs between different fairness notions: using our proposed measures, the overall individual-level unfairness of an algorithm can be decomposed into a between-group and a within-group component. Earlier methods are typically designed to tackle only between-group unfairness, which may be justified for legal or other reasons. However, we demonstrate that minimizing exclusively the between-group component may, in fact, increase the within-group, and hence the overall unfairness. We characterize and illustrate the tradeoffs between our measures of (un)fairness and the prediction accuracy.

12 pages 7 figures To be published in: KDD '18: The 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining Proceedings

References in corpus (7)

• Axiomatic Attribution for Deep Networks
• Equality of Opportunity in Supervised Learning
• Fairness Beyond Disparate Treatment & Disparate Impact: Learning Classification without Disparate Mistreatment
• Inherent Trade-Offs in the Fair Determination of Risk Scores
• Preventing Fairness Gerrymandering: Auditing and Learning for Subgroup Fairness
• Fairness Constraints: Mechanisms for Fair Classification
• Fairness in Learning: Classic and Contextual Bandits

Cited by in corpus (35)

• Fairness in Machine Learning: A Survey
• Fairness in Machine Learning
• Evaluating Stochastic Rankings with Expected Exposure
• Auditing of AI: Legal, Ethical and Technical Approaches
• Fair Preprocessing: Towards Understanding Compositional Fairness of Data Transformers in Machine Learning Pipeline
• Do the Machine Learning Models on a Crowd Sourced Platform Exhibit Bias? An Empirical Study on Model Fairness
• Operationalizing Individual Fairness with Pairwise Fair Representations
• Visual Analysis of Discrimination in Machine Learning
• LiFT: A Scalable Framework for Measuring Fairness in ML Applications
• Mitigating Unfairness via Evolutionary Multi-objective Ensemble Learning
• Towards Individual and Multistakeholder Fairness in Tourism Recommender Systems
• Towards Understanding Fairness and its Composition in Ensemble Machine Learning
• De-biasing "bias" measurement
• What-is and How-to for Fairness in Machine Learning: A Survey, Reflection, and Perspective
• Position is Power: System Prompts as a Mechanism of Bias in Large Language Models (LLMs)
• Algorithmic decision making methods for fair credit scoring
• Crossing the principle-practice gap in AI ethics with ethical problem-solving
• The Fall of an Algorithm: Characterizing the Dynamics Toward Abandonment
• Fairness through Aleatoric Uncertainty
• Accounting for Model Uncertainty in Algorithmic Discrimination
• Loss-Aversively Fair Classification
• Fairness Hacking: The Malicious Practice of Shrouding Unfairness in Algorithms
• Forks Over Knives: Predictive Inconsistency in Criminal Justice Algorithmic Risk Assessment Tools
• Subgroup Fairness in Two-Sided Markets
• AIM: Attributing, Interpreting, Mitigating Data Unfairness
• Formalising Anti-Discrimination Law in Automated Decision Systems
• Group fairness without demographics using social networks
• On the Richness of Calibration
• Fair Inputs and Fair Outputs: The Incompatibility of Fairness in Privacy and Accuracy
• Stairway to Fairness: Connecting Group and Individual Fairness
• Justified Evidence Collection for Argument-based AI Fairness Assurance
• Nishpaksh: TEC Standard-Compliant Framework for Fairness Auditing and Certification of AI Models
• Understanding trade-offs in classifier bias with quality-diversity optimization: an application to talent management
• Profit equitably: An investigation of market maker's impact on equitable outcomes
• Algorithmic Fairness Datasets: the Story so Far