Toward a Theory of Race for Fairness in Machine Learning

Explore a thought-provoking tutorial from the FAT* 2019 conference that delves into the critical intersection of race theory and fairness in machine learning. Examine how computer scientists approach minimizing disparate impacts across racial categories in algorithmic systems, while questioning the underlying concept of race itself. Gain insights into critical race theory and social scientific discourses, and learn how these concepts can be translated for machine learning practitioners. Participate in small-group activities that demonstrate the relevance of these theories to fairness problems in AI. Note that due to technical issues, video images are unavailable for the first 2:34 minutes, but audio is fully accessible throughout the 43-minute presentation. Access the accompanying slides for a comprehensive understanding of this important topic in ethical AI development.

FAT* 2019 Translation Tutorial: Toward a Theory of Race for Fairness in Machine Learning