Nash Equilibrium
Nash equilibrium, a cornerstone of game theory, describes a stable state in a game where no player can improve their outcome by unilaterally changing their strategy. Current research focuses on developing efficient algorithms, such as those based on reinforcement learning, online learning, and Gaussian processes, to compute Nash equilibria in increasingly complex scenarios, including multi-agent systems, Markov games, and games with incomplete information. These advancements are crucial for addressing challenges in diverse fields like robotics, resource allocation, and cybersecurity, where understanding strategic interactions between agents is paramount for designing effective and robust systems.
Papers
Self-Play PSRO: Toward Optimal Populations in Two-Player Zero-Sum Games
Stephen McAleer, JB Lanier, Kevin Wang, Pierre Baldi, Roy Fox, Tuomas Sandholm
A Coupling Approach to Analyzing Games with Dynamic Environments
Brandon C. Collins, Shouhuai Xu, Philip N. Brown
Approximate Nash Equilibrium Learning for n-Player Markov Games in Dynamic Pricing
Larkin Liu
ESCHER: Eschewing Importance Sampling in Games by Computing a History Value Function to Estimate Regret
Stephen McAleer, Gabriele Farina, Marc Lanctot, Tuomas Sandholm
Model-Based Reinforcement Learning for Offline Zero-Sum Markov Games
Yuling Yan, Gen Li, Yuxin Chen, Jianqing Fan
A unified stochastic approximation framework for learning in games
Panayotis Mertikopoulos, Ya-Ping Hsieh, Volkan Cevher