Fitted Q-Learning in Mean-field Games.
Berkay AnahtarciCan Deha KariksizNaci SaldiPublished in: CoRR (2019)
Keyphrases
- reinforcement learning
- cooperative
- function approximation
- multiagent learning
- markov random field
- game theory
- multi agent
- state space
- stochastic approximation
- nash equilibria
- multi agent reinforcement learning
- belief networks
- learning algorithm
- game playing
- game design
- statistical mechanics
- reinforcement learning algorithms
- action selection
- game theoretic
- nash equilibrium
- em algorithm
- bayesian inference
- computer games
- learning agents
- markov decision processes
- dynamic programming
- optimal policy
- linear complexity
- closed form
- video games
- game play
- educational games
- free energy
- variational methods
- stochastic games
- monte carlo
- serious games