Q-Learning in Regularized Mean-field Games.
Berkay AnahtarciCan Deha KariksizNaci SaldiPublished in: Dyn. Games Appl. (2023)
Keyphrases
- reinforcement learning
- cooperative
- multi agent
- multiagent learning
- belief networks
- game theory
- state space
- computer games
- learning algorithm
- markov random field
- statistical mechanics
- function approximation
- video games
- optimal policy
- monte carlo
- policy iteration
- stochastic approximation
- game playing
- game design
- learning agents
- nash equilibria
- game theoretic
- game play
- educational games
- learning rate
- least squares
- closed form
- objective function
- markov networks
- risk minimization
- exact inference
- dynamic programming
- free energy
- multi agent reinforcement learning
- stochastic shortest path
- coalitional games
- stochastic games
- loopy belief propagation
- digital games
- reinforcement learning algorithms
- serious games
- nash equilibrium
- markov decision processes
- mobile robot