Evaluating and Enhancing Equity in AI Systems

Fairlearn: Assessing and Improving Fairness of AI Systems

Hilde Weerts, Miroslav Dudík, Richard Edgar, Adrin Jalali, Roman Lutz, Michael Madaio; 24(257):1−8, 2023.

Abstract

Fairlearn is a project aimed at assisting practitioners in evaluating and enhancing the fairness of artificial intelligence (AI) systems. The fairlearn Python library, which accompanies the project, enables the assessment of a model’s output across different populations and includes multiple algorithms for addressing fairness concerns. Recognizing that fairness is a complex sociotechnical issue, the project incorporates learning resources that help practitioners consider the broader societal implications of a system.

[abs]

[pdf][bib]
      
[code]