On Well-posedness and Minimax Optimal Rates of Nonparametric Q-function Estimation in Off-policy Evaluation.
Xiaohong ChenZhengling QiPublished in: CoRR (2022)
Keyphrases
- policy evaluation
- semi parametric
- worst case
- optimal control
- optimal solution
- monte carlo
- temporal difference
- expected error
- statistical inference
- model free
- dynamic programming
- function approximation
- density estimation
- estimation error
- parametric models
- markov decision processes
- least squares
- reinforcement learning
- evaluation function
- computational complexity