Offline Reinforcement Learning with Differentiable Function Approximation is Provably Efficient.
Ming YinMengdi WangYu-Xiang WangPublished in: CoRR (2022)
Keyphrases
- function approximation
- reinforcement learning
- temporal difference
- temporal difference learning
- model free
- temporal difference learning algorithms
- function approximators
- learning tasks
- radial basis function
- mountain car
- tile coding
- state space
- reinforcement learning algorithms
- learning algorithm
- state action space
- k nearest neighbor
- markov decision processes
- small number
- multi agent
- objective function
- policy gradient
- machine learning