We study the concept of Ex-Post Equilibrium in zero sum games. This solution concept is relevant in games with incomplete information where the payoffs are uncertain, and are assumed to belong to a bounded uncertainty set. A strategy is at Ex-Post Equilibrium if for any payoff matrix in the uncertainty set, the given strategy is a Nash Equilibrium.
We study the existence of such equilibria and show efficient ways to find it for special classes of games.
To read more about our work, check out our paper: https://arxiv.org/pdf/2007.05647.pdf
We present our work at the Theoretical Foundations of Reinforcement Learning Workshop workshop @ICML 2020: https://wensun.github.io/rl_theory_workshop_2020_ICML.github.io/
To see more of our work visit our academic pages:
Wenshuo Guo: https://people.eecs.berkeley.edu/~wguo/
Mihaela Curmei: https://mcurmei627.github.io/
Serena Wang: https://serenalwang.com/
