The goal of my research is to develop a better understanding of how and when machine learning succeeds, and the broader societal implications of data driven systems. My work lies at the intersection of machine learning, optimization, and statistics, including topics such as high dimensional learning and algorithmic fairness.

My recent research spans two lines of work that contribute to these broad themes. My primary line of work builds new theory towards understanding the role of optimization in the success of modern machine learning models, especially deep neural networks. My second line of work is on studying the challenges in building machine learning systems that avoid discrimination against protected population groups.