Learning Nash equilibria with bandit feedback

March 12, 2021, Zoom

Maryam Kamgarpour

U. of British Columbia, Electrical and Computer Engineering


Decision-making in multi-agent systems arises in engineering applications ranging from electricity markets to communication and transportation networks. I discuss decision-making of multiple players with coupled objectives. In this setting, a Nash equilibrium is a stable solution concept, since no agent finds it profitable to unilaterally deviate from her choice. Due to geographic distance, privacy concerns, or simply the scale of these systems, each player can only base her decision on local information. I present our algorithm on learning Nash equilibria in convex games and discuss its convergence.

Speaker's Bio

Maryam Kamgarpour is with the faculty of Electrical and Computer Engineering of the University of British Columbia, Vancouver, Canada. She holds a Doctor of Philosophy in Engineering from the University of California, Berkeley and a Bachelor of Applied Science from University of Waterloo, Canada. Her research is on safe decision-making and control under uncertainty, game theory and mechanism design, mixed integer and stochastic optimization and control. Her theoretical research is motivated by control challenges arising in intelligent transportation networks, robotics, power grid systems and healthcare. She is the recipient of NASA High Potential Individual Award, NASA Excellence in Publication Award, and the European Union (ERC) Starting Grant.
picture: please find attached.

Video URL: