Toward Implicit Learning for the Compositional Verification of Markov Decision Processes.
Redouane BouchekirMohand Cherif BoukalaPublished in: VECoS (2018)
Keyphrases
- markov decision processes
- reinforcement learning
- state space
- dynamic programming
- optimal policy
- model based reinforcement learning
- partially observable
- finite state
- decision theoretic planning
- search algorithm
- real time dynamic programming
- action selection
- reinforcement learning algorithms
- machine learning
- policy iteration
- stochastic games
- planning under uncertainty
- continuous state spaces
- factored mdps
- macro actions
- transition matrices