论文信息 - Linear Programming solvers for Markov Decision Processes

Linear Programming solvers for Markov Decision Processes

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

G. Riano | D. Bello

[1] Shaler Stidham,et al. The Underlying Markov Decision Process in the Single-Leg Airline Yield-Management Problem , 1999, Transp. Sci..

[2] Aparna V. Huzurbazar. Modeling, Analysis, Design, and Control of Stochastic Systems , 2001, Technometrics.

[3] M. Puterman. Chapter 8 Markov decision processes , 1990 .

[4] Martin L. Puterman,et al. Coffee, Tea, or ...?: A Markov Decision Process Model for Airline Meal Provisioning , 2004, Transp. Sci..

[5] R. Faure,et al. Introduction to operations research , 1968 .

[6] P. Samuelson. LIFETIME PORTFOLIO SELECTION BY DYNAMIC STOCHASTIC PROGRAMMING , 1969 .

[7] Martin L. Puterman,et al. Markov Decision Processes: Discrete Stochastic Dynamic Programming , 1994 .

[8] M. Omizo,et al. Modeling , 1983, Encyclopedic Dictionary of Archaeology.

[9] Shaler Stidham,et al. Optimal Control of Markov Chains , 2000 .

[10] George L. Nemhauser,et al. Handbooks in operations research and management science , 1989 .