Abstract
Learning algorithms offer a promising method to solve Nash Equilibrium in games. Meanwhile with the increasing integration of learning algorithms into intelligent and autonomous systems, learning algorithm becomes a agent to play repeated games, leading to emergence of human-machine games. In this talk, we will introduce some work of our team: (1) about the optimal strategy of the human in the human-machine games as well as the periodic/quasi-periodic behavior of the system; (2) about a new NE-solving method based on the asymmetric learning dynamics in zero-sum games; (3) about a new NE-solving method based on the non-convergent dynamics of fictitious play algorithm in non-zero-sum games.
Time
Wednesday, June. 11, 10:00--11:00
Speaker
Dr. Yifen Mu is now an associate professor in Academy of Mathematics and Systems Science, Chinese Academy of Sciences(AMSS, CAS). She received the B.Sc. degree in Mathematics from Peking University, China in 2005, and the Ph.D. degree in System Theory from AMSS, CAS, China, in 2010. Her research interest is game theory and learning in games, especially the human-machine games, learning dynamics in games and learning algorithms for stochastic games. Her co-authored paper “Nash Equilibrium Solving for General Non-Zero-Sum Games from Non-Convergent Dynamics of Fictitious Play” won the Guan Zhao-Zhi Award in 2024(2/1672) . Now she is a council member of Game Theory Branch, Operations Research Society of China (ORSC) an editor board member of Journal of Dynamics and Games.
穆义芬
Room
Room 102
