Non-asymptotic Performances of Robust Markov Decision Processes.
Wenhao YangZhihua ZhangPublished in: CoRR (2021)
Keyphrases
- markov decision processes
- finite state
- state space
- reinforcement learning
- dynamic programming
- transition matrices
- decision theoretic planning
- policy iteration
- optimal policy
- average cost
- action space
- finite horizon
- factored mdps
- reinforcement learning algorithms
- action sets
- planning under uncertainty
- partially observable
- risk sensitive
- reachability analysis
- stochastic shortest path
- decision processes
- infinite horizon
- markov chain
- state abstraction
- supply chain
- convergence rate