Tuesday, April 8
SEC 3.301-3 or Zoom (Passcode: 988031)
11:00am - 12:00pm

Economics-Inspired Models in Machine Learning
Eric Mazumdar, Assistant Professor, Caltech
 

Abstract: Machine learning algorithms are increasingly deployed into environments in which they must interact with other strategic agents like algorithms and people 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 discuss how ideas from economics can give us new insights into the analysis and design of machine learning algorithms for these real-world environments.

In this talk I will show how economics-inspired models can help us overcome fundamental problems in dealing with distribution shifts, other agents, and using black-box ML models for decision-making. First I will discuss a line of work on using models of human decision-making from behavioral economics in multi-agent reinforcement learning. Surprisingly, using these models gives rise to a class of equilibria that can be efficiently computed in all finite-horizon Markov games---allowing us to develop algorithms with strong guarantees of convergence for any multi-agent system. Furthermore, we show that they capture human-play in a variety of game theoretic experiments conducted in behavioral economics. I will then present a line of work on dealing with---and modeling---strategically driven distribution shifts. I will present a simple pde model that captures features of real-world distribution shifts that previous models miss and show how our model can give new insights into the effectiveness of retraining and optimization procedures in machine learning.

Bio: Eric Mazumdar is an Assistant Professor in Computing and Mathematical Sciences and Economics at Caltech. His research lies at the intersection of machine learning and economics where he is broadly interested in developing the tools and understanding necessary to confidently deploy machine learning algorithms into societal-scale systems. Eric is the recipient of a NSF Career Award and was a fellow at the Simons Institute for Theoretical Computer Science for the semester on Learning in Games. He obtained his Ph.D in Electrical Engineering and Computer Science at UC Berkeley where he was advised by Michael Jordan and Shankar Sastry and received his B.S. in Eletrical Engineering and Computer Science at the Massachusetts Institute of Technology (MIT).