Symblicit algorithms for mean-payoff and shortest path in monotonic Markov decision processes.
Aaron BohyVéronique BruyèreJean-François RaskinNathalie BertrandPublished in: Acta Informatica (2017)
Keyphrases
- markov decision processes
- shortest path
- policy iteration
- shortest path algorithm
- factored mdps
- dynamic programming
- reachability analysis
- reinforcement learning
- finite state
- state space
- shortest path problem
- decision theoretic planning
- optimal policy
- strongly connected components
- shortest distance
- optimal path
- transition matrices
- model free
- markov decision problems
- partially observable markov decision processes
- machine learning
- game theory
- optimization problems
- multi agent
- learning algorithm