@misc{10498/32822,
year = {2024},
url = {http://hdl.handle.net/10498/32822},
abstract = {Bias in data may lead to prediction procedures which discriminate individuals from sensitive groups. In this paper we propose a Bayesian method for parameter estimation in the linear regression model taking fairness into ac- count. For a given prior structure, namely, the Normal-Gamma prior, for a specific choice of unfairness measure (which does not require the use of private -sensitive- information), our method computes a constrained empirical Bayes estimator that induces fairness in posterior predictions. Specifically, the model hyper-parameters are calculated as the optimal solution of a constrained optimization problem: that in which the marginal likelihood is maximized in a region where the unfairness measure is upper bounded and thus, controlled. This guarantees that average posterior predictions are, at most, as unfair as the upper bound establishes. Experi- ments with synthetic and real data sets show the competitiveness of our approach in terms of fairness in the obtained solution and prediction error.},
keywords = {fairness},
keywords = {machine learning},
keywords = {linear regression model},
keywords = {Normal-Gamma prior,},
keywords = {empirical Bayes},
keywords = {unfairness measure},
keywords = {Gram-Schmidt process},
title = {Empirical Bayes fairness in linear regression},
author = {Carrizosa, Emilio and Jiménez-Llamas, Rafael and Ramírez Cobo, Josefa},
}