A Fast Analytical Algorithm for Solving Markov Decision Processes with Real-Valued Resources.
Janusz MareckiSven KoenigMilind TambePublished in: IJCAI (2007)
Keyphrases
- real valued
- markov decision processes
- dynamic programming
- learning algorithm
- model based reinforcement learning
- reinforcement learning
- np hard
- expectation maximization
- policy iteration
- average reward
- transition matrices
- state space
- least squares
- probabilistic planning
- optimal policy
- semi markov decision processes
- action space
- model free
- probabilistic model
- optimal solution
- training data