Belangrijkste concepten
This paper introduces a novel learning algorithm, Independent Stochastic Policy-Nested-Gradient (ISPNG), that enables agents to efficiently learn approximate Nash equilibria in adversarial team Markov games (ATMGs) using only individual rewards and state observations as feedback.
Kalogiannis, F., Yan, J., & Panageas, I. (2024). Learning Equilibria in Adversarial Team Markov Games: A Nonconvex-Hidden-Concave Min-Max Optimization Problem. arXiv preprint arXiv:2410.05673.
This paper addresses the open problem of efficiently learning Nash equilibria in adversarial team Markov games (ATMGs) where agents have limited information and communication.