On the Existence of Easy Initial States for Undiscounted Stochastic Games.
S. H. TijsO. J. VriezePublished in: Math. Oper. Res. (1986)
Keyphrases
- stochastic games
- markov decision processes
- nash equilibria
- multi agent
- multiagent reinforcement learning
- learning automata
- average reward
- state transitions
- nash equilibrium
- reinforcement learning algorithms
- repeated games
- robust optimization
- imperfect information
- finite state machines
- game theoretic
- initial state
- cooperative
- reinforcement learning