Coco-Q: Learning in Stochastic Games with Side Payments.
Eric SodomkaElizabeth HilliardMichael L. LittmanAmy GreenwaldPublished in: ICML (3) (2013)
Keyphrases
- stochastic games
- reinforcement learning algorithms
- state action
- multi agent reinforcement learning
- multi agent
- reinforcement learning
- single agent
- learning agent
- markov decision processes
- rl algorithms
- state space
- model free
- multiagent reinforcement learning
- nash equilibria
- average reward
- learning algorithm
- cooperative
- function approximation
- imperfect information
- optimal policy
- nash equilibrium
- temporal difference
- reward function
- policy iteration
- solving problems
- repeated games
- dynamic environments
- multiple agents
- infinite horizon
- robust optimization
- decision problems
- least squares
- multi agent systems
- action selection
- genetic algorithm