Memory-two strategies forming symmetric mutual reinforcement learning equilibrium in repeated prisoners' dilemma game.
Masahiko UedaPublished in: Appl. Math. Comput. (2023)
Keyphrases
- mixed strategy
- reinforcement learning
- nash equilibrium
- repeated games
- equilibrium strategies
- game theory
- pure strategy
- nash equilibria
- games with incomplete information
- solution concepts
- optimal strategy
- stochastic games
- markov games
- reinforcement learning algorithms
- game theoretic
- incomplete information
- subgame perfect equilibrium
- evolutionary game theory
- equilibrium point
- learning agents
- exploration exploitation dilemma
- function approximation
- learning algorithm
- variational inequalities
- computer games
- exploration strategy
- multi agent reinforcement learning
- two player games
- correlated equilibrium
- extensive form games
- temporal difference learning
- multi agent
- fictitious play
- temporal difference
- optimal control
- markov decision processes
- perfect information
- dynamic programming
- minimax search
- game playing
- cooperative
- virtual world