← List all talks ...

Challenges & Opportunities for Learning in Strategic Environments

Eric Mazumdar – Assistant Professor, Caltech

Thu, 25-Apr-2024 / 4:00pm / Packard 202

Abstract

Machine learning algorithms are increasingly being deployed into environments in which they must interact with other strategic agents with potentially misaligned objectives. While the presence of these strategic interactions creates new challenges for learning algorithms, they also give rise to new opportunities for algorithm design.

In this talk, I will begin by highlighting a line of work showing the unintuitive behaviors that arise from the interplay between learning algorithms and strategic agents. First, I will show—both in theory and practice—how learning algorithms (even those built on top of large language models) are susceptible to the gaming of data by vanishingly small groups of people, even if individuals have no effect on them in isolation. I will also present recent work on how strategic interactions can break our basic intuition that larger models, more data, and more compute always improves performance. The resulting Braess paradox-like phenomenon suggests that—even when one has access to infinite data—strategic interactions can make smaller and less expressive models yield better equilibrium outcomes. I will conclude with some recent work on algorithm design in strategic environments in the context of multi-agent RL. In particular, I will show how to tweak deep Q learning to allow it to have strong convergence guarantees in competitive games.

Bio

Eric Mazumdar is an Assistant Professor in Computing and Mathematical Sciences and Economics at Caltech. He obtained his Ph.D in Electrical Engineering and Computer Science at UC Berkeley, co-advised by Michael Jordan and Shankar Sastry.

His research interests lie at the intersection of machine learning and economics. He is broadly interested in developing the tools and understanding necessary to confidently deploy machine learning algorithms into societal systems. This requires understanding the theoretical underpinnings of learning algorithms in uncertain, dynamic environments where they interact with strategic agents— including people and other algorithms. Practically, He applies his work to problems in intelligent infrastructure, online markets, e-commerce, and the delivery of healthcare.

He is the recipient of a NSF Career Award aimed at studying the strategic interactions that arise in Societal-Scale Systems as well as a Research Fellowship for Learning in Games from the Simons Institute for Theoretical Computer Science. His work is supported by NSF, DARPA, and Amazon research grants.

Prior to Berkeley, he received an SB in Electrical Engineering and Computer Science at Massachusetts Institute of Technology (MIT), where he had the opportunity to work with in the Laboratory for Multiscale Regenerative Technologies as well as in the MIT Computational Biology Group in CSAIL.