Rényi Fair Inference

Abstract

Machine learning algorithms have been increasingly deployed in critical automated decision-making systems that directly affect human lives. When these algorithms are only trained to minimize the training/test error, they could suffer from systematic discrimination against individuals based on their sensitive attributes such as gender or race. Recently, there has been a surge in machine learning society to develop algorithms for fair machine learning. In particular, many adversarial learning procedures have been proposed to impose fairness. Unfortunately, these algorithms either can only impose fairness up to first-order dependence between the variables, or they lack computational convergence guarantees. In this paper, we use Rényi correlation as a measure of fairness of machine learning models and develop a general training framework to impose fairness. In particular, we propose a min-max formulation which balances the accuracy and fairness when solved to optimality. For the case of discrete sensitive attributes, we suggest an iterative algorithm with theoretical convergence guarantee for solving the proposed min-max problem. Our algorithm and analysis are then specialized to fair classification and the fair clustering problem under disparate impact doctrine. Finally, the performance of the proposed Rényi fair inference framework is evaluated on Adult and Bank datasets.

11 pages, 1 figure

References in corpus (14)

• Equality of Opportunity in Supervised Learning
• Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddings
• Preventing Fairness Gerrymandering: Auditing and Learning for Subgroup Fairness
• A Convex Framework for Fair Regression
• Learning Adversarially Fair and Transferable Representations
• Solving a Class of Non-Convex Min-Max Games Using Iterative First Order Methods
• A Reductions Approach to Fair Classification
• What is Local Optimality in Nonconvex-Nonconcave Minimax Optimization?
• Learning Controllable Fair Representations
• On the Convergence and Robustness of Training GANs with Regularized Optimal Transport
• Fairness for Robust Log Loss Classification
• Fair Coresets and Streaming Algorithms for Fair k-Means Clustering
• Unleashing Linear Optimizers for Group-Fair Learning and Optimization
• Minimum HGR Correlation Principle: From Marginals to Joint Distribution

Cited by in corpus (7)

• Non-convex Min-Max Optimization: Applications, Challenges, and Recent Theoretical Advances
• Individual Fairness for $k$-Clustering
• Fairness-Aware Neural Réyni Minimization for Continuous Features
• A Max-Min Entropy Framework for Reinforcement Learning
• Nonconvex-Nonconcave Min-Max Optimization with a Small Maximization Domain
• Fairness Through Regularization for Learning to Rank
• Fair for All: Best-effort Fairness Guarantees for Classification