A Monte Carlo Neural Fictitious Self-Play approach to approximate Nash Equilibrium in imperfect-information dynamic games.
Li ZhangYuxuan ChenWei WangZiliang HanShijian LiZhijie PanGang PanPublished in: Frontiers Comput. Sci. (2021)
Keyphrases
- imperfect information
- monte carlo
- nash equilibrium
- game theoretic
- game tree search
- game theory
- game tree
- fictitious play
- opponent modeling
- perfect information
- nash equilibria
- stochastic games
- game playing
- markov chain
- card game
- importance sampling
- solution concepts
- imperfect information games
- monte carlo tree search
- particle filter
- mixed strategy
- dynamic environments
- worst case
- cooperative
- sufficient conditions
- human players
- stackelberg game
- optimal strategy
- lower bound
- multi agent systems
- pure nash equilibrium
- decision making