Structural Estimation of Markov Decision Processes in High-Dimensional State Space with Finite-Time Guarantees.
Siliang ZengMingyi HongAlfredo GarciaPublished in: CoRR (2022)
Keyphrases
- markov decision processes
- state and action spaces
- optimal policy
- state space
- dynamic programming
- finite state
- transition matrices
- reinforcement learning
- policy iteration
- decision theoretic planning
- action space
- average cost
- high dimensional state space
- partially observable
- average reward
- planning under uncertainty
- markov decision process
- infinite horizon
- reachability analysis
- reward function
- model based reinforcement learning
- stationary policies
- action sets
- graph cuts
- stochastic shortest path