Model-based reinforcement learning for on-line feedback-Nash equilibrium solution of N-player nonzero-sum differential games.
Rushikesh KamalapurkarJustin KlotzWarren E. DixonPublished in: ACC (2014)
Keyphrases
- nash equilibrium
- repeated games
- pure nash equilibria
- game theory
- game theoretic
- nash equilibria
- imperfect information
- variational inequalities
- model based reinforcement learning
- solution concepts
- stackelberg game
- fictitious play
- pareto optimal
- mixed strategy
- pure strategy
- congestion games
- normal form games
- incomplete information
- stochastic games
- equilibrium strategies
- optimal solution
- pure nash equilibrium
- objective function
- markov decision processes
- worst case
- special case
- learning algorithm