Decentralized Q-learning for weakly acyclic stochastic dynamic games.
Gürdal ArslanSerdar YükselPublished in: CDC (2015)
Keyphrases
- stochastic dynamic
- multi agent
- cooperative
- game theory
- reinforcement learning
- multiagent learning
- function approximation
- computer games
- video games
- game design
- game theoretic
- np complete
- state space
- np hard
- optimal policy
- learning agents
- stochastic approximation
- temporal difference learning
- game playing
- nash equilibria
- learning algorithm
- multi agent reinforcement learning
- helmholtz principle
- reinforcement learning algorithms
- stochastic games
- action selection
- game play
- educational games
- distributed systems
- coalitional games
- serious games
- nash equilibrium
- action sets
- dynamic programming