Eric Mazumdar

Eric Mazumdar is an Assistant Professor of Computing and Mathematical Sciences and Economics at the California Institute of Technology (Caltech), holding dual appointments in the Division of Engineering and Applied Science and the Humanities and Social Sciences division. His research bridges machine learning and economics with applications in societal-scale systems.

His educational background includes a B.S. in Electrical Engineering and Computer Science from MIT (2015) and a Ph.D. in Electrical Engineering and Computer Science from UC Berkeley (2021), where he was co-advised by Michael Jordan and Shankar Sastry.

Mazumdar's research focuses on theoretical foundations of learning algorithms in uncertain, dynamic environments interacting with strategic agents. Key areas include algorithmic game theory, reinforcement learning in strategic settings, risk-aware decision-making, and behavioral economics. His work addresses challenges in intelligent infrastructure, online markets, e-commerce, and healthcare delivery through rigorous mathematical frameworks.

His publication trends reveal a strong focus on zero-sum stochastic games, distributionally robust reinforcement learning, and strategic classification, with increasing integration of optimization theory and game-theoretic principles. Recent work emphasizes sample efficiency, risk aversion, and equilibrium computation in multi-agent systems.

• NSF Career Award for studying strategic interactions in Societal-Scale Systems
• Research Fellowship for Learning in Games from the Simons Institute

Mazumdar advises multiple postdocs (Zaiwei Chen, Laixi Shi, Kishan Panaganti) and graduate students (Lauren Conger, Tinashe Handina, Yizhou Zhang), with research funded by NSF, DARPA, and Amazon. His group explores applications in intelligent infrastructure and healthcare systems through collaborations with Caltech's Linde Institute and Center for Social Information Sciences (CSIS).