ECE SEMINAR

Speaker: Max Fishelson, Massachusetts Institute of Technology
Title: Theoretical Foundations for Multi-Agent Learning
Day: Monday, February 16, 2026
Time: 4:30 PM
Location: B205 Engineering Quadrangle
Host: Pramod Viswanath

Abstract:
As artificial intelligence becomes ubiquitous in complex, interactive systems, we need models that perform reliably under dynamic conditions. AI systems are moving beyond passive perception to active participation: controlling autonomous vehicles in mixed traffic, participating in markets, and managing decentralized networks. In these roles, an agent's decisions reshape the environment and invite strategic responses from others. This feedback loop invalidates a core assumption of classical machine learning: that data is independent and identically distributed. We require a new foundation for learning that is inherently robust to these adaptive, multi-agent dynamics.

In this talk, I will demonstrate how we can approach this challenge using regret minimization and calibration. Instead of measuring performance against a static distribution, these measures allow us to derive provable guarantees against any sequence of outcomes. An algorithm with low regret effectively “learns to play” optimally, even against strategic counter-parties. Calibration guarantees that we can distill trustworthy forecasts from dynamic data. I will discuss my work pushing the boundaries of what is possible in this domain, resolving long-standing open questions on regret minimization in games, the scalability of algorithms for strong notions of regret, and the fundamental limits of trustworthy forecasting. Together, these results provide an algorithmic pathway toward AI systems that remain robust and reliable even in complex, interactive environments.

Bio:
Max Fishelson is a final-year PhD student in the Theory of Computation group at MIT CSAIL, advised by Constantinos Daskalakis. His research focuses on the theoretical foundations of machine learning, specifically at the intersection of learning theory and algorithmic game theory. He works on designing algorithms with provable guarantees for decision-making in adversarial and multi-agent environments.