Achieving the Asymptotically Minimax Optimal Sample Complexity of Offline Reinforcement Learning: A DRO-Based Approach.
Yue WangJinjun XiongShaofeng ZouPublished in: Trans. Mach. Learn. Res. (2024)
Keyphrases
- sample complexity
- reinforcement learning
- worst case
- upper bound
- learning problems
- sequential decision problems
- sample size
- learning algorithm
- vc dimension
- supervised learning
- theoretical analysis
- pac learning
- dynamic programming
- lower bound
- active learning
- optimal control
- special case
- generalization error
- state space
- optimal solution
- markov decision processes
- machine learning
- transfer learning
- feature extraction
- partially observable