Model-free adaptive dynamic programming for online optimal solution of the unknown nonlinear zero-sum differential game.
Chunbin QinHuaguang ZhangYanhong LuoPublished in: IJCNN (2014)
Keyphrases
- model free
- dynamic programming
- reinforcement learning
- stochastic games
- optimal solution
- game theory
- average reward
- reinforcement learning algorithms
- initially unknown
- repeated games
- function approximation
- policy iteration
- nash equilibria
- nash equilibrium
- boolean games
- np hard
- game playing
- temporal difference
- linear programming
- optimal policy
- state space
- markov decision processes
- search space
- objective function
- optimal control
- perfect information
- policy evaluation
- incomplete information
- computer games
- infinite horizon
- neural network