√n-Regret for Learning in Markov Decision Processes with Function Approximation and Low Bellman Rank.
Kefan DongJian PengYining WangYuan ZhouPublished in: CoRR (2019)
Keyphrases
- reinforcement learning
- function approximation
- markov decision processes
- temporal difference learning
- reinforcement learning algorithms
- learning tasks
- learning algorithm
- function approximators
- partially observable
- actor critic
- optimal policy
- model free
- state action
- stochastic games
- state space
- total reward
- markov decision process
- supervised learning
- learning process
- finite state
- real time dynamic programming
- temporal difference
- machine learning
- action selection
- monte carlo
- reinforcement learning methods
- artificial neural networks
- data mining