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: 4:10PM 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 email@example.com, with “join” in the subject line and a blank email body. All talks info and reminders will be sent via the mailing list.
Title: Control Theoretic Modeling of Bounded Rationality in Games
Speaker: Jeff Shamma
Date and Time: 12/02/2021 4:10PM ET
Location: Phillips 233 and Zoom
Abstract: The classical solution concept of Nash equilibrium in game theory presumes that each agent’s decision is optimal with respect to the decisions of others. Bounded rationality modeling, for which there is a long history and extensive body of research, seeks to understand game outcomes where the capacity of agents to act as optimizers is limited, as would be the case in large-scale or complex environments. This talk presents two approaches to bounded rationality inspired by concepts in feedback control. The first is in the setting of learning in games, where agents utilize simple adaptive rules over the course of repeated interactions to adjust their decisions. We present an overview of the control theoretic concept of passivity, which captures the behavior of many learning rules considered in the learning in games literature. This model imposes an intuitive dynamic constraint on how decisions evolve in reaction to dynamic rewards. We present preliminary experimental results that examine the degree to which participant decisions are consistent with passivity modeling. The second is in the setting of partially observed stochastic games. We describe an alternative equilibrium concept, termed empirical evidence equilibria, where agents utilize reduced order models of both the environment and decision rules of other agents. We establish existence results for such equilibria and, in special settings, compare to Nash equilibria, correlated equilibria, and mean field games.
Bio: Jeff S. Shamma is with the University of Illinois at Urbana-Champaign where he is the Department Head of Industrial and Enterprise Systems Engineering (ISE) and Jerry S. Dobrovolny Chair in ISE. His prior academic appointments include faculty positions at the King Abdullah University of Science and Technology (KAUST) and Georgia Institute of Technology, where he was the Julian T. Hightower Chair in Systems and Controls. Jeff received a PhD in Systems Science and Engineering from MIT in 1988. He is a Fellow of IEEE and IFAC; a recipient of the IFAC High Impact Paper Award, AACC Donald P. Eckman Award, and NSF Young Investigator Award; and a past Distinguished Lecturer of the IEEE Control Systems Society. Jeff is currently serving as Editor-in-Chief for the IEEE Transactions on Control of Network Systems.
Schedule for Fall 2021:
|9/9/2021||Alon Orlisky||Robust Learning from Batches: The Best Things in life are (Almost) Free|
|9/23/2021||Qing Zhao||Random Walk on a Tree for Stochastic Search and Optimization|
|10/7/2021||Mary Wootters||Low-bandwidth recovery of linear functions of Reed-Solomon-encoded data|
|10/21/2021||Anna Scaglione||A User Guide to Low-Pass Graph Signal Processing and its Applications|
|11/04/2021||Steven H. Low||Optimal Power Flow and Smart EV Charging|
|11/18/2021||Vikram Krishnamurthy||Adversarial Filtering and Inverse Reinforcement Learning|
|12/02/2021||Jeff Shamma||Control Theoretic Modeling of Bounded Rationality in Games|