Stackelberg Game

Stackelberg games model strategic interactions where a leader commits to a strategy, influencing a follower's best response. Current research focuses on extending the framework to handle uncertainties, incomplete information (e.g., about follower objectives), and decentralized learning scenarios, often employing reinforcement learning, game-theoretic algorithms, and optimization techniques to find optimal or near-optimal strategies for both leader and follower. These advancements are improving the applicability of Stackelberg games to diverse fields, including security, resource allocation, and multi-agent systems, by providing robust and efficient solutions for complex decision-making problems under uncertainty. The development of algorithms that handle non-myopic followers and address distributional robustness are particularly significant recent contributions.

Papers

February 5, 2022

Adversarially Trained Actor Critic for Offline Reinforcement Learning
Ching-An Cheng, Tengyang Xie, Nan Jiang, Alekh Agarwal
Offline Reinforcement Learning Deep RL Stackelberg Game Public Policy Offline RL Algorithm

February 3, 2022

Stackelberg Strategic Guidance for Heterogeneous Robots Collaboration
Yuhan Zhao, Baichuan Huang, Jingjin Yu, Quanyan Zhu
Game Theory Heterogeneous Robot Stackelberg Game Stackelberg Equilibrium Coordination Strategy Stackelberg Planning

Stackelberg Game

Papers

Adversarially Trained Actor Critic for Offline Reinforcement Learning

Stackelberg Strategic Guidance for Heterogeneous Robots Collaboration