Demonic variance and a non-determinism score for Markov decision processes.
Jakob PiribauerPublished in: CoRR (2024)
Keyphrases
- markov decision processes
- reinforcement learning
- optimal policy
- state space
- dynamic programming
- finite state
- policy iteration
- transition matrices
- finite horizon
- partially observable
- infinite horizon
- factored mdps
- reinforcement learning algorithms
- decision processes
- average cost
- decision theoretic planning
- model based reinforcement learning
- action space
- planning under uncertainty
- policy evaluation
- state abstraction
- reachability analysis
- average reward
- risk sensitive
- markov decision process
- action sets
- partially observable markov decision processes
- interval estimation
- state and action spaces
- objective function