A statistical framework for fair predictive algorithms

Abstract

Predictive modeling is increasingly being employed to assist human decision-makers. One purported advantage of replacing human judgment with computer models in high stakes settings-- such as sentencing, hiring, policing, college admissions, and parole decisions-- is the perceived "neutrality" of computers. It is argued that because computer models do not hold personal prejudice, the predictions they produce will be equally free from prejudice. There is growing recognition that employing algorithms does not remove the potential for bias, and can even amplify it, since training data were inevitably generated by a process that is itself biased. In this paper, we provide a probabilistic definition of algorithmic bias. We propose a method to remove bias from predictive models by removing all information regarding protected variables from the permitted training data. Unlike previous work in this area, our framework is general enough to accommodate arbitrary data types, e.g. binary, continuous, etc. Motivated by models currently in use in the criminal justice system that inform decisions on pre-trial release and paroling, we apply our proposed method to a dataset on the criminal histories of individuals at the time of sentencing to produce "race-neutral" predictions of re-arrest. In the process, we demonstrate that the most common approach to creating "race-neutral" models-- omitting race as a covariate-- still results in racially disparate predictions. We then demonstrate that the application of our proposed method to these data removes racial disparities from predictions with minimal impact on predictive accuracy.

Cited by in corpus (27)

• Fairness in Machine Learning: A Survey
• Fairness in Machine Learning
• Beyond Parity: Fairness Objectives for Collaborative Filtering
• Achieving Fairness through Adversarial Learning: an Application to Recidivism Prediction
• Empirical Risk Minimization under Fairness Constraints
• Noise-tolerant fair classification
• Inherent Tradeoffs in Learning Fair Representations
• Leveraging Labeled and Unlabeled Data for Consistent Fair Binary Classification
• Improving the Fairness of Deep Generative Models without Retraining
• The Authority of "Fair" in Machine Learning
• Tuning Fairness by Balancing Target Labels
• Fairness guarantee in multi-class classification
• Removing biased data to improve fairness and accuracy
• A Near-Optimal Algorithm for Debiasing Trained Machine Learning Models
• Debiasing representations by removing unwanted variation due to protected attributes
• Predicting Court Decisions for Alimony: Avoiding Extra-legal Factors in Decision made by Judges and Not Understandable AI Models
• Technologies for Trustworthy Machine Learning: A Survey in a Socio-Technical Context
• An Information-theoretic Approach to Distribution Shifts
• Decomposition of Global Feature Importance into Direct and Associative Components (DEDACT)
• Towards a Measure of Individual Fairness for Deep Learning
• Fairness in Ranking under Uncertainty
• Fair Classification via Unconstrained Optimization
• A Fully Bayesian, Logistic Regression Tracking Algorithm for Mitigating Disparate Misclassification
• An example of prediction which complies with Demographic Parity and equalizes group-wise risks in the context of regression
• Towards Auditability for Fairness in Deep Learning
• Bias-Tolerant Fair Classification
• Attraction-Repulsion clustering with applications to fairness