Finding Safe Zones of policies Markov Decision Processes.
Lee CohenYishay MansourMichal MoshkovitzPublished in: CoRR (2022)
Keyphrases
- markov decision processes
- optimal policy
- markov decision process
- decision processes
- average cost
- state space
- finite horizon
- decentralized control
- reward function
- reinforcement learning
- policy iteration
- dynamic programming
- discounted reward
- transition matrices
- average reward
- total reward
- infinite horizon
- finite state
- macro actions
- reinforcement learning algorithms
- partially observable markov decision processes
- decision problems
- factored mdps
- stationary policies
- expected reward
- policy iteration algorithm
- control policies
- planning under uncertainty
- decision theoretic planning
- partially observable
- reachability analysis
- model based reinforcement learning
- state and action spaces
- action space
- action sets
- semi markov decision processes
- long run
- multistage
- multiagent systems