Gizem Soğancıoğlu

Using Explainability for Bias Mitigation: A Case Study for Fair Recruitment Assessment

2023 · INTERNATIONAL CONFERENCE ON MULTIMODAL INTERACTION

In this study, we propose a bias-mitigation algorithm, dubbed ProxyMute, that uses an explainability method to detect proxy features of a given sensitive attribute (e.g., gender) and reduces their effects on decisions by disabling them …