Near-optimal Policy Identification in Active Reinforcement Learning.
Xiang LiViraj MehtaJohannes KirschnerIan CharWillie NeiswangerJeff SchneiderAndreas KrauseIlija BogunovicPublished in: ICLR (2023)
Keyphrases
- optimal policy
- reinforcement learning
- markov decision processes
- state space
- finite horizon
- decision problems
- long run
- dynamic programming
- multistage
- average reward
- finite state
- markov decision process
- state dependent
- function approximation
- infinite horizon
- average cost
- policy iteration
- sufficient conditions
- bayesian reinforcement learning
- initial state
- multi agent
- serial inventory systems
- lost sales
- model free
- control policies
- markov decision problems
- total reward
- reward function