Offline Policy Evaluation in Large Action Spaces via Outcome-Oriented Action Grouping.
Jie PengHao ZouJiashuo LiuShaoming LiYibao JiangJian PeiPeng CuiPublished in: WWW (2023)
Keyphrases
- action space
- policy evaluation
- markov decision processes
- reinforcement learning
- policy iteration
- state space
- temporal difference
- action selection
- model free
- markov decision problems
- real valued
- optimal policy
- least squares
- dynamic programming
- function approximation
- state action
- continuous action
- finite state
- infinite horizon
- average reward
- monte carlo
- reinforcement learning algorithms
- average cost
- stochastic processes
- markov decision process
- function approximators
- optimal control
- machine learning
- partially observable
- markov chain
- learning algorithm