Non-asymptotic Convergence of Adam-type Reinforcement Learning Algorithms under Markovian Sampling.
Huaqing XiongTengyu XuYingbin LiangWei ZhangPublished in: CoRR (2020)
Keyphrases
- reinforcement learning algorithms
- asymptotic convergence
- reinforcement learning
- state space
- model free
- markov decision processes
- reinforcement learning problems
- function approximation
- learning algorithm
- reinforcement learning methods
- temporal difference
- dynamic environments
- monte carlo
- fixed point
- dynamic programming
- stochastic games
- sample size
- reward function