Title :
Linear Programming solvers for Markov Decision Processes
Author :
Bello, Diego ; Riano, German
Author_Institution :
Univ. de Los Andes
Abstract :
This paper describes linear programming solvers for Markov decision processes, as an extension to the JMDP program. JMDP is an object-oriented framework to model and solve Markov decision processes (MDP) programmed in Java. The developed solvers work for the discounted cost and average cost criteria. Our solvers are compared with existing value iteration and policy iteration solvers
Keywords :
Java; Markov processes; decision theory; iterative methods; linear programming; mathematics computing; JMDP program; Java; Markov decision processes; linear programming solvers; object-oriented framework; policy iteration solvers; value iteration solvers; Costs; Design engineering; Infinite horizon; Inventory management; Java; Linear programming; Log periodic antennas; Object oriented modeling; Optimal control; Systems engineering and theory;
Conference_Titel :
Systems and Information Engineering Design Symposium, 2006 IEEE
Conference_Location :
Charlottesville, VA
Print_ISBN :
1-4244-0474-6
Electronic_ISBN :
1-4244-0474-6
DOI :
10.1109/SIEDS.2006.278719
