Learning Non-Discriminatory Predictors

Abstract

We consider learning a predictor which is non-discriminatory with respect to a "protected attribute" according to the notion of "equalized odds" proposed by Hardt et al. [2016]. We study the problem of learning such a non-discriminatory predictor from a finite training set, both statistically and computationally. We show that a post-hoc correction approach, as suggested by Hardt et al, can be highly suboptimal, present a nearly-optimal statistical procedure, argue that the associated computational problem is intractable, and suggest a second moment relaxation of the non-discrimination definition for which learning is tractable.

28 pages

Cited by in corpus (61)

• Fairness of Exposure in Rankings
• Fairness in Machine Learning: A Survey
• User-oriented Fairness in Recommendation
• A Convex Framework for Fair Regression
• Fairness in Credit Scoring: Assessment, Implementation and Profit Implications
• Personalized Counterfactual Fairness in Recommendation
• A Reductions Approach to Fair Classification
• Empirical Risk Minimization under Fairness Constraints
• Fairness in Risk Assessment Instruments: Post-Processing to Achieve Counterfactual Equalized Odds
• Fairness for Robust Log Loss Classification
• Robust Optimization for Fairness with Noisy Protected Groups
• Marrying Fairness and Explainability in Supervised Learning
• Unbiased Graph Embedding with Biased Graph Observations
• Classification with Fairness Constraints: A Meta-Algorithm with Provable Guarantees
• Noise-tolerant fair classification
• Improved Adversarial Learning for Fair Classification
• Fair Learning with Private Demographic Data
• Promoting Fairness through Hyperparameter Optimization
• End-To-End Bias Mitigation: Removing Gender Bias in Deep Learning
• Improving the Fairness of Deep Generative Models without Retraining
• Recent advances in deep learning theory
• Constrained Learning with Non-Convex Losses
• DeBayes: a Bayesian Method for Debiasing Network Embeddings
• Review of Mathematical frameworks for Fairness in Machine Learning
• Training Fair Models in Federated Learning without Data Privacy Infringement
• Equalized odds postprocessing under imperfect group information
• Auditing and Achieving Intersectional Fairness in Classification Problems
• Equalizing Recourse across Groups
• Tuning Fairness by Balancing Target Labels
• Envy-Free Classification
• Capuchin: Causal Database Repair for Algorithmic Fairness
• Rényi Fair Inference
• Learning Fair Scoring Functions: Bipartite Ranking under ROC-based Fairness Constraints
• Wasserstein-based fairness interpretability framework for machine learning models
• Removing biased data to improve fairness and accuracy
• Fairness With Minimal Harm: A Pareto-Optimal Approach For Healthcare
• Bias, Fairness, and Accountability with AI and ML Algorithms
• Predicting Early Dropout: Calibration and Algorithmic Fairness Considerations
• The stochastic multi-gradient algorithm for multi-objective optimization and its application to supervised machine learning
• Unified Group Fairness on Federated Learning
• Fairness in Reinforcement Learning
• Statistical Equity: A Fairness Classification Objective
• A Near-Optimal Algorithm for Debiasing Trained Machine Learning Models
• Fairness-aware Agnostic Federated Learning
• To Split or Not to Split: The Impact of Disparate Treatment in Classification
• Adaptive Sampling for Minimax Fair Classification
• Fair Classification and Social Welfare
• Technologies for Trustworthy Machine Learning: A Survey in a Socio-Technical Context
• Fairness Through Regularization for Learning to Rank
• User Acceptance of Gender Stereotypes in Automated Career Recommendations
• Quantifying Infra-Marginality and Its Trade-off with Group Fairness
• Metadata Normalization
• FairMod - Making Predictive Models Discrimination Aware
• Towards a Measure of Individual Fairness for Deep Learning
• Fair quantile regression
• GroupMixNorm Layer for Learning Fair Models
• Fair Classification with Adversarial Perturbations
• Learning from Discriminatory Training Data
• Attraction-Repulsion clustering with applications to fairness
• Towards Auditability for Fairness in Deep Learning
• Global Optimum Search in Quantum Deep Learning