Events

Multi-agent reinforcement learning for collaborative games: a network and mean-field perspective

Electrical and Computer Engineering
Finance and Risk Engineering
Lecture / Panel
 
Open to the Public
""

Speaker

Renyuan Xu
Assistant professor, Department of Finance and Risk Engineering, New York University.

Title

"System Noise and Individual Exploration in Learning Large Population Games"

Abstract

Multi-agent reinforcement learning (MARL) has enjoyed substantial successes in many applications including real-time resource allocation, order matching for ride-hailing, and autonomous driving. Despite the empirical success of MARL, general theories behind MARL algorithms are less developed due to the intractability of interactions, complex information structure, and the curse of dimensionality. Instead of directly analyzing the multi-agent systems, the mean-field theory provides a powerful approach to approximate the games under various notions of equilibria. Moreover, the analytical feasible framework of mean-field theory leads to efficient and tractable learning algorithms with theoretical guarantees.

In this talk, we will demonstrate how mean-field theory can contribute to analyzing a class of simultaneous-learning-and-decision-making problems under cooperation, with unknown rewards and dynamics. Moreover, we will show that the learning procedure can be further decentralized and scaled up if a network structure is specified. Our result lays the first theoretical foundation for the so-called "centralized training and decentralized execution" scheme, a widely used training scheme in the empirical works of cooperative MARL problems.
 

About Speaker

I am currently an assistant professor at the Department of Finance and Risk Engineering at New York University. Before joining NYU, I was an assistant professor in the Daniel J. Epstein Department of Industrial and Systems Engineering at the University of Southern California from 2021-2024, and a Hooke Research Fellow in the Mathematical Institute at the University of Oxford from 2019-2021. I completed my Ph.D. in 2019 at the University of California, Berkeley in the Department of Industrial Engineering and Operations Research. 

My research interests include stochastic analysis, stochastic controls and games, machine learning theory, and mathematical finance. I am also interested in interdisciplinary topics that integrate methodologies in multiple fields such as applied probability, statistics, and optimization, along with their applications in addressing high-stake decision-making problems in modern large-scale systems. Some of the topics that I have been working on recently:

  • Stochastic games and mean-field games with applications in finance
  • Generative AI and deep learning theory through the lens of stochastic control and differential equations
  • Reinforcement learning theory and applications in algorithmic trading 
  • Stochastic controls under imperfect observations and dynamic information acquisition
Zoom
Speakers