Stochastic Stability of Reinforcement Learning in Positive-Utility Games.
Georgios C. ChasparisPublished in: CoRR (2017)
Keyphrases
- reinforcement learning
- direct policy search
- action sets
- stochastic approximation
- learning agents
- learning automata
- function approximation
- state space
- control policies
- reinforcement learning agents
- monte carlo
- utility function
- positive and negative
- stochastic programming problems
- optimal policy
- continuous state spaces
- game theory
- multi agent
- nash equilibrium
- computer games
- learning algorithm
- multiagent learning
- stability analysis
- game playing
- game play
- video games
- transferable utility
- nash equilibria
- game theoretic
- markov decision processes
- game players
- temporal difference learning
- control system
- control policy
- temporal difference
- optimal control