Avoiding Discrimination through Causal Reasoning

Abstract

Recent work on fairness in machine learning has focused on various statistical discrimination criteria and how they trade off. Most of these criteria are observational: They depend only on the joint distribution of predictor, protected attribute, features, and outcome. While convenient to work with, observational criteria have severe inherent limitations that prevent them from resolving matters of fairness conclusively. Going beyond observational criteria, we frame the problem of discrimination based on protected attributes in the language of causal reasoning. This viewpoint shifts attention from "What is the right fairness criterion?" to "What do we want to assume about the causal data generating process?" Through the lens of causality, we make several contributions. First, we crisply articulate why and when observational criteria fail, thus formalizing what was before a matter of opinion. Second, our approach exposes previously ignored subtleties and why they are fundamental to the problem. Finally, we put forward natural causal non-discrimination criteria and develop algorithms that satisfy them.

Advances in Neural Information Processing Systems 30, 2017 http://papers.nips.cc/paper/6668-avoiding-discrimination-through-causal-reasoning

References in corpus (3)

• On the (im)possibility of fairness
• Exposing the Probabilistic Causal Structure of Discrimination
• Fairness in Criminal Justice Risk Assessments: The State of the Art

Cited by in corpus (30)

• Prediction-Based Decisions and Fairness: A Catalogue of Choices, Assumptions, and Definitions
• Fairness of Exposure in Rankings
• Bias and Discrimination in AI: a cross-disciplinary perspective
• A Clarification of the Nuances in the Fairness Metrics Landscape
• An Empirical Characterization of Fair Machine Learning For Clinical Risk Prediction
• Personalized Counterfactual Fairness in Recommendation
• The Disparate Effects of Strategic Manipulation
• Fairness in Algorithmic Decision Making: An Excursion Through the Lens of Causality
• Visual Analysis of Discrimination in Machine Learning
• Why Fairness Cannot Be Automated: Bridging the Gap Between EU Non-Discrimination Law and AI
• Fairness in Risk Assessment Instruments: Post-Processing to Achieve Counterfactual Equalized Odds
• The Impact of Explanations on Layperson Trust in Artificial Intelligence-Driven Symptom Checker Apps: Experimental Study
• Simpson's paradox in Covid-19 case fatality rates: a mediation analysis of age-related causal effects
• Amazon SageMaker Clarify: Machine Learning Bias Detection and Explainability in the Cloud
• Bias In, Bias Out? Evaluating the Folk Wisdom
• A General Framework for Fair Regression
• Toward Operationalizing Pipeline-aware ML Fairness: A Research Agenda for Developing Practical Guidelines and Tools
• BeFair: Addressing Fairness in the Banking Sector
• The Life Cycle of Knowledge in Big Language Models: A Survey
• Demographic Bias: A Challenge for Fingervein Recognition Systems?
• Regulatory Instruments for Fair Personalized Pricing
• A Human-Centric Perspective on Fairness and Transparency in Algorithmic Decision-Making
• Decoupled classifiers for fair and efficient machine learning
• Fairness in Forecasting of Observations of Linear Dynamical Systems
• Fairness Through Counterfactual Utilities
• Counterfactual Situation Testing: From Single to Multidimensional Discrimination
• MobilityMirror: Bias-Adjusted Transportation Datasets
• On the Direction of Discrimination: An Information-Theoretic Analysis of Disparate Impact in Machine Learning
• GroupMixNorm Layer for Learning Fair Models
• Questioning causality on sex, gender and COVID-19, and identifying bias in large-scale data-driven analyses: the Bias Priority Recommendations and Bias Catalog for Pandemics