Constrained discounted Markov decision processes with Borel state spaces.
Eugene A. FeinbergAnna JaskiewiczAndrzej S. NowakPublished in: Autom. (2020)
Keyphrases
- markov decision processes
- state space
- optimal policy
- reinforcement learning algorithms
- dynamic programming
- reinforcement learning
- average cost
- finite state
- markov decision process
- finite horizon
- infinite horizon
- policy iteration
- average reward
- action space
- partially observable
- factored mdps
- model based reinforcement learning
- markov chain
- decision processes
- transition matrices
- planning problems
- planning under uncertainty
- stationary policies
- discounted reward
- reachability analysis
- state abstraction
- heuristic search
- search space
- long run
- dynamical systems
- semi markov decision processes
- reward function
- state variables
- evaluation function
- action sets