Discrete-Time Deterministic Q-Learning: A Novel Convergence Analysis.
Qinglai WeiFrank L. LewisQiuye SunPengfei YanRuizhuo SongPublished in: IEEE Trans. Cybern. (2017)
Keyphrases
- convergence analysis
- global convergence
- function approximation
- reinforcement learning
- cooperative
- state space
- optimality conditions
- learning algorithm
- multi agent
- markov chain
- convergence rate
- learning rate
- reinforcement learning algorithms
- monte carlo
- global optimum
- dynamic programming
- optimal policy
- model free
- simulated annealing
- reward function