Actor-Critic Fictitious Play in Simultaneous Move Multistage Games.
Julien PérolatBilal PiotOlivier PietquinPublished in: AISTATS (2018)
Keyphrases
- multistage
- fictitious play
- actor critic
- game theory
- nash equilibria
- approximate dynamic programming
- dynamic programming
- reinforcement learning
- optimal control
- nash equilibrium
- policy gradient
- temporal difference
- neuro fuzzy
- imperfect information
- gradient method
- lot sizing
- optimal policy
- policy iteration
- reinforcement learning algorithms
- cooperative
- incomplete information
- average cost
- evaluation function
- game theoretic
- linear program
- stochastic games
- function approximation
- monte carlo
- model free
- resource allocation
- cost function