The FIND Seminar is a bi-weekly seminar series that hosts cutting-edge research talks on topics related to the broad themes of Foundations of Information, Networks and Decision Systems. Talks are about 50 minutes long with time for questions and discussion.
Location: Phillips Hall 233 and Zoom
Time: 4:15PM ET, bi-weekly on (alternating) Thursdays
Delivery format: All talks will have a live audience in Phillips Hall 233. Until circumstances allow otherwise, external speakers will give the talk remotely via Zoom (broadcasted in PH233). Remote audience is also welcome, but in-person participation is encouraged.
Mailing list: To subscribe to the FIND seminar mailing list, email find-seminar-l-request@cornell.edu, with “join” in the subject line and a blank email body. All talks info and reminders will be sent via the mailing list.
Upcoming Talk
Title: Multi-Agent Reinforcement Learning in Markov Potential Games and Beyond
Speaker: Manxi Wu
Date and Time: 10/05/2023 4:15PM ET
Location: Phillips Hall 233 and Zoom

Abstract:
Infinite-horizon stochastic games provide a versatile framework for studying the repeated interaction among multiple strategic agents in dynamic environments. However, computing equilibria in such games is highly complex, and the long-run outcomes of decentralized learning algorithms in multi-agent settings remain poorly understood. The first part of this talk introduces a multi-agent reinforcement learning dynamics tailored for independent and decentralized settings, where players lack knowledge of the game model and cannot coordinate. The proposed dynamics guarantee convergence to a stationary Nash equilibrium in Markov potential games, demonstrating the effectiveness of simple learning dynamics even with limited information. In the second part of the talk, we extend the learning framework to encompass Markov near potential games, offering flexibility to incorporate a wide range of practically-relevant multi-agent interaction settings. We present efficient algorithms for approximating the stationary Nash equilibrium and substantiate their effectiveness through regret analysis and numerical experiments.
Schedule for Fall 2023:
Date | Speaker | Title |
08/24/2023 | Amit Prakash | Building Analytics Tools for everyone using LLMs |
09/07/2023 | Itzhak Tamo | Nonlinear Repair Schemes of Reed-Solomon Codes |
09/21/2023 | FIND Community | FIND Graduate Day |
09/27/2023 | Robert Nowak | Cancelled. |
10/05/2023 | Manxi Wu | Multi-Agent Reinforcement Learning in Markov Potential Games and Beyond |
10/19/2023 | John Simpson-Porco | TBD |
11/02/2023 | Enrique Mallada | TBD |
11/16/2023 | Panayotis Mertikopoulos | TBD |
11/30/2023 | Johannes Ballé | TBD |
A list of previous talks can be found here.