Closing the Gender Wage Gap: Adversarial Fairness in Job Recommendation
This addresses the gender wage gap problem for job seekers by providing a method to reduce algorithmic bias in recommendation systems, though it is incremental as it builds on existing adversarial debiasing techniques.
The study tackled gender bias in job recommendations by using a generative adversarial network to debias word2vec representations from 12M job vacancy texts and 900k resumes, resulting in the elimination of a wage gap where women were previously recommended lower-salary jobs.
The goal of this work is to help mitigate the already existing gender wage gap by supplying unbiased job recommendations based on resumes from job seekers. We employ a generative adversarial network to remove gender bias from word2vec representations of 12M job vacancy texts and 900k resumes. Our results show that representations created from recruitment texts contain algorithmic bias and that this bias results in real-world consequences for recommendation systems. Without controlling for bias, women are recommended jobs with significantly lower salary in our data. With adversarially fair representations, this wage gap disappears, meaning that our debiased job recommendations reduce wage discrimination. We conclude that adversarial debiasing of word representations can increase real-world fairness of systems and thus may be part of the solution for creating fairness-aware recommendation systems.