Hort, Max; (2023) Investigating Trade-offs For Fair Machine Learning Systems. Doctoral thesis (Ph.D), UCL (University College London).

Preview

Text Thesis-Max_Hort.pdf - Other Download (7MB) | Preview

Abstract

Fairness in software systems aims to provide algorithms that operate in a nondiscriminatory manner, with respect to protected attributes such as gender, race, or age. Ensuring fairness is a crucial non-functional property of data-driven Machine Learning systems. Several approaches (i.e., bias mitigation methods) have been proposed in the literature to reduce bias of Machine Learning systems. However, this often comes hand in hand with performance deterioration. Therefore, this thesis addresses trade-offs that practitioners face when debiasing Machine Learning systems. At first, we perform a literature review to investigate the current state of the art for debiasing Machine Learning systems. This includes an overview of existing debiasing techniques and how they are evaluated (e.g., how is bias measured). As a second contribution, we propose a benchmarking approach that allows for an evaluation and comparison of bias mitigation methods and their trade-offs (i.e., how much performance is sacrificed for improving fairness). Afterwards, we propose a debiasing method ourselves, which modifies already trained Machine Learning models, with the goal to improve both, their fairness and accuracy. Moreover, this thesis addresses the challenge of how to deal with fairness with regards to age. This question is answered with an empirical evaluation on real-world datasets.

Download activity - last month

Export as JSONXMLCSV

Download activity - last 12 months

Export as JSONXMLCSV

Downloads by country - last 12 months

Export as JSONXMLCSV

Archive Staff Only

View Item