Monte Carlo Tree Search for Verifying Reachability in Markov Decision Processes.
Pranav AshokTomás BrázdilJan KretínskýOndrej SlámeckaPublished in: CoRR (2018)
Keyphrases
- markov decision processes
- monte carlo tree search
- state space
- monte carlo
- reinforcement learning
- optimal policy
- markov chain
- finite state
- reinforcement learning methods
- evaluation function
- reinforcement learning algorithms
- policy iteration
- temporal difference
- decision theoretic planning
- particle filter
- action space
- temporal difference learning
- dynamic programming
- transition matrices
- model checking
- heuristic search
- partially observable
- markov decision process
- real time dynamic programming
- game tree
- average reward
- infinite horizon
- initial state
- average cost
- state variables
- belief state
- stochastic shortest path
- partially observable markov decision processes
- planning problems
- learning experience
- computational complexity