Mitigating Bias in Algorithmic Systems -- A Fish-Eye View

TL;DR

This paper provides a broad overview of strategies to identify, manage, and explain bias in algorithmic systems, emphasizing the importance of stakeholder perspectives and cross-domain approaches.

Contribution

It offers a comprehensive survey of bias mitigation methods across four research areas, highlighting the three key steps: detection, fairness management, and explainability.

Findings

Identifies three main steps in bias mitigation: detection, fairness management, explainability.

Emphasizes the importance of stakeholder perspectives in addressing bias.

Highlights the need for integrated, system-wide approaches to bias mitigation.

Abstract

Mitigating bias in algorithmic systems is a critical issue drawing attention across communities within the information and computer sciences. Given the complexity of the problem and the involvement of multiple stakeholders -- including developers, end-users, and third parties -- there is a need to understand the landscape of the sources of bias, and the solutions being proposed to address them, from a broad, cross-domain perspective. This survey provides a "fish-eye view," examining approaches across four areas of research. The literature describes three steps toward a comprehensive treatment -- bias detection, fairness management and explainability management -- and underscores the need to work from within the system as well as from the perspective of stakeholders in the broader context.