Model-based RL with Optimistic Posterior Sampling: Structural Conditions and Sample Complexity.
Alekh AgarwalTong ZhangPublished in: NeurIPS (2022)
Keyphrases
- sample complexity
- sample size
- theoretical analysis
- learning algorithm
- pac learning
- learning problems
- reinforcement learning
- vc dimension
- model free
- active learning
- upper bound
- special case
- generalization error
- supervised learning
- random sampling
- sufficient conditions
- training examples
- probability distribution
- lower bound
- concept classes
- multi agent
- markov decision processes
- state space
- gaussian process
- irrelevant features
- training set
- prior knowledge
- kernel function
- dynamic programming