Stability of the Nash equilibrium under gradient ascent learning algorithms in two-agent two-action games.
Amit BhayaRodrigo Brandolt Sodre de MacedoLucas Shiguemitsu ShigueokaPublished in: CCA (2013)
Keyphrases
- nash equilibrium
- gradient ascent
- stochastic games
- state action
- learning algorithm
- game theory
- nash equilibria
- policy gradient
- action selection
- game theoretic
- stackelberg game
- multi agent systems
- normal form games
- solution concepts
- learning agent
- multiagent systems
- fictitious play
- mixed strategy
- reinforcement learning
- pure strategy
- worst case
- exponential family
- cross entropy
- expectation maximization
- machine learning
- multi agent
- regret minimization
- pure nash equilibria
- repeated games
- average reward
- dynamic environments
- multiple agents
- autonomous agents
- evaluation function
- imperfect information
- markov decision processes
- learning capabilities
- reward function
- maximum likelihood
- cooperative
- action space
- incomplete information