Algorithmic Game Theory

Computing equilibria and learning strategies in multi-agent strategic settings.

Many real-world systems involve strategic interactions — defenders and attackers, planners and responders. We develop scalable methods for computing and learning equilibria in these settings.

Key directions:

  • End-to-end gradient descent for Stackelberg games
  • Deep fictitious play for continuous action spaces
  • Security games and green security patrol scheduling
  • Learning adversary models from data

Selected publications:

(Wang et al., 2022; Niu et al., 2025; Kamra et al., 2019; Cooney et al., 2019; Wang et al., 2020)

References

2025

  1. Finding a Multiple Follower Stackelberg Equilibrium: A Fully First-Order Method
    April Niu, Kai Wang, and Juba Ziani
    2025
    In submission
    arXiv

2022

  1. Coordinating Followers to Reach Better Equilibria: End-to-End Gradient Descent for Stackelberg Games
    Kai Wang, Lily Xu, Andrew Perrault, Michael K. Reiter, and Milind Tambe
    In AAAI Conference on Artificial Intelligence (AAAI), 2022
    arXiv

2020

  1. Scalable Game-Focused Learning of Adversary Models: Data-to-Decisions in Network Security Games
    Kai Wang, Andrew Perrault, Aditya Mate, and Milind Tambe
    In International Conference on Autonomous Agents and Multi-Agent Systems (AAMAS), 2020
    HTML

2019

  1. DeepFP for Finding Approximate Nash Equilibrium in Continuous Action Spaces
    Nitin Kamra, Umang Gupta, Kai Wang, Fei Fang, Yan Liu, and 1 more author
    In Conference on Decision and Game Theory for Security (GameSec), 2019
    HTML
  2. Learning to Signal in the Goldilocks Zone: Improving Adversary Compliance in Security Games
    Sarah Cooney, Kai Wang, Elizabeth Bondi, Thanh Nguyen, Phebe Vayanos, and 5 more authors
    In European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML-PKDD), 2019
    HTML