Racial Categories in Machine Learning
The Co-Opting AI series and Race and Public Space Working Group at NYU’s Institute for Public Knowledge invite you to join for a discussion of Racial Categories in Machine Learning, with Sebastian Benthall and Bruce D. Haynes.
Controversies around race and machine learning have sparked debate among computer scientists over how to design machine learning systems that guarantee fairness. These debates rarely engage with how racial identity is embedded in our social experience, making for sociological and psychological complexity. This complexity challenges the paradigm of considering fairness to be a formal property of supervised learning with respect to protected personal attributes. Racial identity is not simply a personal subjective quality. For people labeled “Black” it is an ascribed political category that has consequences for social differentiation embedded in systemic patterns of social inequality achieved through both social and spatial segregation. In the United States, racial classification can best be understood as a system of inherently unequal status categories that places whites as the most privileged category while signifying the Negro/black category as stigmatized. Social stigma is reinforced through the unequal distribution of societal rewards and goods along racial lines that is reinforced by state, corporate, and civic institutions and practices. This creates a dilemma for society and designers: be blind to racial group disparities and thereby reify racialized social inequality by no longer measuring systemic inequality, or be conscious of racial categories in a way that itself reifies race. We propose a third option. By preceding group fairness interventions with unsupervised learning to dynamically detect patterns of segregation, machine learning systems can mitigate the root cause of social disparities, social segregation and stratification, without further anchoring status categories of disadvantage.
Sebastian Benthall is a security scientist working at the intersection of computer science, economics, law, and philosophy. He is a Research Scholar at NYU’s Information Law Institute and Center for Cybersecurity. His current interests are around compliance engineering and data economics. He has worked at the Digital Life Initiative at Cornell Tech. Before becoming a scientist, Sebastian managed the development of spatial data infrastructure for global coordination around disaster risk reduction. He holds a B.A. in Cognitive Science from Brown University and is completing his PhD at UC Berkeley’s School of Information.
Bruce D. Haynes is professor of sociology at the University of California, Davis and a Senior Fellow in the Urban Ethnography Project at Yale University. His publications include The Ghetto: Contemporary Issues and Controversies (co-author R. Hutchison, Westview Press, 2011), Red Lines, Black Spaces: The Politics of Race and Space in a Black Middle-Class Suburb (Yale University Press 2001 reissued in paper in 2006), and Down the Up Staircase: Three Generations of a Harlem Family (co-author Syma Solovitch, Columbia University Press 2017). Down the Up Staircase tells the story of one Harlem family across three generations, connecting its journey to the historical and social forces that transformed Harlem over the past century. Haynes and Solovitch capture the tides of change that pushed blacks forward through the twentieth century—the Great Migration, the Harlem Renaissance, the early civil rights victories, the Black Power and Black Arts movements—as well as the many forces that ravaged black communities, including Haynes’s own.
Image: 1938 Home Owners’ Loan Corporation map of Brooklyn, from the National Archives and Records Administration, Mapping Inequality.