Infinite-Horizon Offline Reinforcement Learning with Linear Function Approximation: Curse of Dimensionality and Algorithm.
Lin ChenBruno ScherrerPeter L. BartlettPublished in: CoRR (2021)
Keyphrases
- function approximation
- reinforcement learning
- model free
- mountain car
- function approximators
- dynamic programming
- infinite horizon
- learning algorithm
- temporal difference learning
- policy iteration
- temporal difference
- optimal control
- monte carlo
- temporal difference learning algorithms
- markov decision processes
- optimal policy
- search space
- objective function
- cost function
- reinforcement learning algorithms
- mathematical model
- sufficient conditions
- probability distribution
- partially observable
- optimal solution
- actor critic
- decision making
- data mining