Folding Algorithm for Policy Evaluation for Markov Decision Processes With Quasi-Birth Death Structure.
Yassir YassirLangford B. WhitePublished in: IEEE Trans. Autom. Control. (2015)
Keyphrases
- markov decision processes
- policy evaluation
- policy iteration
- dynamic programming
- reinforcement learning
- cost function
- learning algorithm
- model free
- computational complexity
- monte carlo
- state space
- least squares
- optimal policy
- optimal solution
- finite state
- average reward
- search algorithm
- linear programming
- convergence rate
- multi agent