Finding Safe Zones of Markov Decision Processes Policies.
Lee CohenYishay MansourMichal MoshkovitzPublished in: NeurIPS (2023)
Keyphrases
- markov decision processes
- optimal policy
- markov decision process
- decision processes
- state space
- reward function
- average cost
- finite state
- total reward
- decentralized control
- discounted reward
- transition matrices
- dynamic programming
- expected reward
- partially observable markov decision processes
- reinforcement learning
- policy iteration algorithm
- finite horizon
- decision problems
- macro actions
- reinforcement learning algorithms
- planning under uncertainty
- average reward
- infinite horizon
- policy iteration
- reachability analysis
- markov decision problems
- decision theoretic planning
- state and action spaces
- multistage
- control policies
- stationary policies
- partially observable
- semi markov decision processes
- risk sensitive
- model based reinforcement learning
- linear programming
- action space
- multi agent
- markov chain
- sufficient conditions
- long run