Review of Mathematical frameworks for Fairness in Machine Learning

Abstract

A review of the main fairness definitions and fair learning methodologies proposed in the literature over the last years is presented from a mathematical point of view. Following our independence-based approach, we consider how to build fair algorithms and the consequences on the degradation of their performance compared to the possibly unfair case. This corresponds to the price for fairness given by the criteria $\text{statistical parity}$ or $\text{equality of odds}$. Novel results giving the expressions of the optimal fair classifier and the optimal fair predictor (under a linear regression gaussian model) in the sense of $\text{equality of odds}$ are presented.

arXiv admin note: substantial text overlap with arXiv:2001.07864, arXiv:1911.04322, arXiv:1906.05082 by other authors

References in corpus (8)

• Equality of Opportunity in Supervised Learning
• Data Decisions and Theoretical Implications when Adversarially Learning Fair Representations
• A Convex Framework for Fair Regression
• From Parity to Preference-based Notions of Fairness in Classification
• On the relation between accuracy and fairness in binary classification
• Fair Regression: Quantitative Definitions and Reduction-based Algorithms
• Kernel Dependence Regularizers and Gaussian Processes with Applications to Algorithmic Fairness
• Fast Fair Regression via Efficient Approximations of Mutual Information

Cited by in corpus (1)

• Classification with abstention but without disparities