Improved Sample Complexity for Reward-free Reinforcement Learning under Low-rank MDPs.
Yuan ChengRuiquan HuangJing YangYingbin LiangPublished in: CoRR (2023)
Keyphrases
- reinforcement learning
- low rank
- sample complexity
- learning problems
- learning algorithm
- supervised learning
- markov decision processes
- convex optimization
- missing data
- matrix factorization
- theoretical analysis
- linear combination
- function approximation
- vc dimension
- state space
- upper bound
- reward function
- singular value decomposition
- generalization error
- high dimensional data
- semi supervised
- reinforcement learning algorithms
- temporal difference
- kernel matrix
- lower bound
- high order
- optimal policy
- active learning
- machine learning
- training examples
- model free
- special case
- partially observable
- machine learning algorithms
- markov decision process
- transfer learning
- model selection
- learning process
- feature selection
- data sets
- markov decision problems
- semi supervised learning
- principal component analysis