Offline RL with Observation Histories: Analyzing and Improving Sample Complexity.
Joey HongAnca D. DraganSergey LevinePublished in: ICLR (2024)
Keyphrases
- sample complexity
- learning algorithm
- theoretical analysis
- learning problems
- reinforcement learning
- supervised learning
- upper bound
- pac learning
- vc dimension
- active learning
- special case
- lower bound
- generalization error
- pac learnability
- concept classes
- training examples
- sample size
- sample complexity bounds
- markov decision processes
- optimal policy
- state space
- irrelevant features
- uniform convergence
- linear threshold
- covering numbers
- prior knowledge
- decision trees