Adaptive Sampling for Best Policy Identification in Markov Decision Processes.
Aymen Al MarjaniAlexandre ProutièrePublished in: ICML (2021)
Keyphrases
- markov decision processes
- adaptive sampling
- optimal policy
- policy iteration
- markov decision process
- average reward
- average cost
- infinite horizon
- partially observable
- monte carlo
- state and action spaces
- reward function
- finite horizon
- state space
- finite state
- decision processes
- action space
- reinforcement learning
- transition matrices
- random sampling
- dynamic programming
- decision problems
- policy evaluation
- total reward
- discounted reward
- reinforcement learning algorithms
- long run
- sufficient conditions
- planning under uncertainty
- partially observable markov decision processes
- markov decision problems
- expected reward
- decision theoretic planning
- stationary policies
- continuous state spaces
- least squares
- control charts
- artificial neural networks