Feb 5, 2021 04:00 PM Athens via Zoom
Probably Approximately Correct Nash Equilibrium Learning
In this talk we consider a multi-agent noncooperative game with agents’ objective functions being affected by uncertainty. Following a data driven paradigm, we represent uncertainty by means of scenarios and seek a robust Nash equilibrium solution. We treat the Nash equilibrium computation problem within the realm of probably approximately correct (PAC) learning. Building upon recent developments in scenario-based optimization, we accompany the computed Nash equilibrium with a priori and a posteriori probabilistic robustness certificates, providing confidence that the computed equilibrium remains unaffected (in probabilistic terms) when a new uncertainty realization is encountered. We demonstrate the efficacy of our approach on the problem of optimal charging in electric vehicle charging control games.
About the Speaker
Kostas Margellos received the Diploma in electrical engineering from the University of Patras, Greece, in 2008, and the Ph.D. in control engineering from ETH Zurich, Switzerland, in 2012. He spent 2013, 2014 and 2015 as a postdoctoral researcher at ETH Zurich, UC Berkeley and Politecnico di Milano, respectively. In 2016 he joined the Control Group, Department of Engineering Science, University of Oxford, where he is currently an Associate Professor. He is also a Fellow at Reuben College and a Lecturer at Worcester College.
In 2018-19 and 2019-20 he received a Gold Teaching Excellence Award, Department of Engineering Science, University of Oxford. He is currently serving as an Associate Editor for Automatica. His research interests include optimization, control and learning complex uncertain systems, with applications to energy networks and shared mobility systems.
You can download the slides of the presentation