I am a Computing Ph.D. student at the University of Utah, where I am advised by Suresh Venkatasubramanian. I am interested in analyzing the social impact of black-box machine learning systems to develop better AI law and policy. My research has been supported by the ARCS Foundation.
email: kumari at cs dot utah dot edu
Shapley Residuals: Quantifying the limits of the Shapley value for explanations. I. Elizabeth Kumar, Carlos Scheidegger, Suresh Venkatasubramanian, Sorelle Friedler. Presented at the 5th ICML Workshop on Human Interpretability in Machine Learning (WHI), 2020.
Problems with Shapley-value-based explanations as feature importance measures. I. Elizabeth Kumar, Suresh Venkatasubramanian, Carlos Scheidegger, Sorelle Friedler. Forthcoming in Proceedings of the 37th International Conference on Machine Learning (ICML), 2020.