Fairness in Supervised Learning: An Information Theoretic Approach

TL;DR

This paper introduces an information theoretic framework for creating fair supervised learning models that prevent discrimination based on sensitive attributes by using data compression and equalized odds.

Contribution

It proposes a novel method that compresses data into an auxiliary variable to ensure fairness and generalization, addressing bias in decision-making systems.

Findings

Framework effectively reduces discrimination in predictions.

Auxiliary variable ensures fairness without sacrificing accuracy.

Method generalizes well across different datasets.

Abstract

Automated decision making systems are increasingly being used in real-world applications. In these systems for the most part, the decision rules are derived by minimizing the training error on the available historical data. Therefore, if there is a bias related to a sensitive attribute such as gender, race, religion, etc. in the data, say, due to cultural/historical discriminatory practices against a certain demographic, the system could continue discrimination in decisions by including the said bias in its decision rule. We present an information theoretic framework for designing fair predictors from data, which aim to prevent discrimination against a specified sensitive attribute in a supervised learning setting. We use equalized odds as the criterion for discrimination, which demands that the prediction should be independent of the protected attribute conditioned on the actual label. To ensure fairness and generalization simultaneously, we compress the data to an auxiliary variable, which is used for the prediction task. This auxiliary variable is chosen such that it is decontaminated from the discriminatory attribute in the sense of equalized odds. The final predictor is obtained by applying a Bayesian decision rule to the auxiliary variable.