Safe Policy Improvement Approaches on Discrete Markov Decision Processes.
Philipp SchollFelix DietrichClemens OtteSteffen UdluftPublished in: CoRR (2022)
Keyphrases
- markov decision processes
- optimal policy
- policy iteration
- markov decision process
- state space
- finite horizon
- average cost
- continuous state spaces
- average reward
- partially observable
- infinite horizon
- decision processes
- dynamic programming
- finite state
- action space
- decision theoretic planning
- reward function
- reinforcement learning
- policy evaluation
- transition matrices
- partially observable markov decision processes
- decision problems
- state and action spaces
- model based reinforcement learning
- planning under uncertainty
- factored mdps
- discounted reward
- markov decision problems
- expected reward
- reachability analysis
- state abstraction
- multistage