Fair machine learning: Influence of demographic parity fairness on other fairness measures

Abstract

Machine learning is used more and more every day. In machine learning, fairness is a continuously growing field of interest with an increasing number of projects every year. Fairness is involved in machine learning to remove biased outcomes, or at least make it less biased. When making use of fairness, there is often a focus on one fairness definition. In this paper, the influence of Zafar et al.’s implementation (Zafar, 2018) of demographic parity on other fairness measures will be discussed.