Fitted Value Iteration in Continuous MDPs With State Dependent Action Sets.
Hao LiShiping ShaoAbhishek GuptaPublished in: IEEE Control. Syst. Lett. (2022)
Keyphrases
- markov decision processes
- state dependent
- action sets
- optimal policy
- action space
- finite state
- state space
- reinforcement learning
- decision problems
- finite horizon
- continuous state
- policy iteration
- average cost
- dynamic programming
- stationary policies
- markov decision process
- long run
- average reward
- initial state
- sufficient conditions
- multistage
- steady state
- asymptotically optimal
- decision processes
- infinite horizon
- partially observable
- markov chain
- single server
- reward function
- partially observable markov decision processes
- stationary distribution
- arrival rate
- queueing networks
- multi agent
- model free
- heuristic search
- learning algorithm