Online Bootstrap Inference For Policy Evaluation in Reinforcement Learning.
Pratik RamprasadYuantong LiZhuoran YangZhaoran WangWill Wei SunGuang ChengPublished in: CoRR (2021)
Keyphrases
- policy evaluation
- reinforcement learning
- temporal difference
- least squares
- model free
- function approximation
- policy iteration
- monte carlo
- markov decision processes
- statistical inference
- td learning
- optimal policy
- state space
- variance reduction
- dynamic programming
- reinforcement learning algorithms
- semi parametric
- evaluation function
- bayesian networks
- bayesian inference
- multi agent
- decision problems
- step size
- cross validation
- partially observable
- regression model
- markov decision problems
- upper bound