Juan Perdomo Silva

Juan Perdomo Silva is an Assistant Professor in the Department of Computer Science at New York University. His research focuses on AI & Society, Machine Learning Theory, and Algorithms. Starting in Fall 2026, he will be joining NYU as an Assistant Professor of Computer Science and Data Science. His work explores the feedback loops between people and algorithms, with an emphasis on developing methods with provable guarantees in dynamic, social contexts. He earned his PhD in EECS from UC Berkeley and holds a BA in CS and Math from Harvard. Currently, he is a postdoc at MIT after serving as a CRCS postdoctoral fellow at Harvard.

Juan Carlos Perdomo is a postdoctoral researcher at MIT, focusing on machine learning and optimization. His research explores the intersection of prediction and decision-making in dynamic systems, with an emphasis on performative prediction, where predictions influence future outcomes. He investigates methods to stabilize dynamical systems, optimize performative risk, and improve fairness in algorithmic decision-making. His work also addresses challenges in transfer learning, policy evaluation, and robustness against adversarial attacks. Perdomo's publications highlight the importance of predictive models in social systems, including education and public policy, while emphasizing the need for stable and private estimation techniques in machine learning.