Keywords: [ ReScience - MLRC 2021 ] [ Journal Track ]
The presented study evaluates ''Exacerbating Algorithmic Bias through Fairness Attacks'' by Mehrabi et al. (2021) within the scope of the ML Reproducibility Challenge 2021. We find it not possible to reproduce the original results from sole use of the paper, and difficult even in possession of the provided codebase. Yet, we managed to obtain similar findings that supported three out of the five main claims of the publication, albeit using partial re-implementations and numerous assumptions. On top of the reproducibility study, we also extend the work of the authors by implementing a different stopping method, which changes the effectiveness of the proposed attacks.