On the Sample Complexity of Vanilla Model-Based Offline Reinforcement Learning with Dependent Samples.
Mustafa O. KarabagUfuk TopcuPublished in: AAAI (2023)
Keyphrases
- reinforcement learning
- model free
- data sets
- function approximation
- state space
- real time
- sample points
- statistically independent
- robotic control
- reinforcement learning algorithms
- optimal policy
- training samples
- learning process
- machine learning
- data mining
- real world
- supervised learning
- search space
- multi agent
- case study
- optimal control
- knowledge base
- learning algorithm
- learning capabilities
- action space
- temporal difference learning
- spectral data