Information Theoretic Measures for Fairness-aware Feature Selection

TL;DR

This paper introduces an information theoretic framework for fairness-aware feature selection that accounts for feature correlations and decision impacts to mitigate bias in machine learning models.

Contribution

It proposes novel information theoretic measures and a fairness utility score for feature selection, considering data joint statistics rather than specific classifiers.

Findings

Effective bias measurement using information theoretic metrics

Feature importance assessed via Shapley value analysis

Framework validated on real and synthetic datasets

Abstract

Machine learning algorithms are increasingly used for consequential decision making regarding individuals based on their relevant features. Features that are relevant for accurate decisions may however lead to either explicit or implicit forms of discrimination against unprivileged groups, such as those of certain race or gender. This happens due to existing biases in the training data, which are often replicated or even exacerbated by the learning algorithm. Identifying and measuring these biases at the data level is a challenging problem due to the interdependence among the features, and the decision outcome. In this work, we develop a framework for fairness-aware feature selection which takes into account the correlation among the features and the decision outcome, and is based on information theoretic measures for the accuracy and discriminatory impacts of features. In particular, we first propose information theoretic measures which quantify the impact of different subsets of features on the accuracy and discrimination of the decision outcomes. We then deduce the marginal impact of each feature using Shapley value function; a solution concept in cooperative game theory used to estimate marginal contributions of players in a coalitional game. Finally, we design a fairness utility score for each feature (for feature selection) which quantifies how this feature influences accurate as well as nondiscriminatory decisions. Our framework depends on the joint statistics of the data rather than a particular classifier design. We examine our proposed framework on real and synthetic data to evaluate its performance.