Linear programming solvers for Markov Decision Processes

被引：12
|
作者：
Bello, Diego [1 ]
Riano, German [1 ]
机构：
[1] Univ Los Andes, COPA, Bogota, Colombia
来源：
2006 IEEE SYSTEMS AND INFORMATION ENGINEERING DESIGN SYMPOSIUM | 2006年
关键词：
D O I：
10.1109/SIEDS.2006.278719
中图分类号：
TP18 [人工智能理论];
学科分类号：
081104 ; 0812 ; 0835 ; 1405 ;
摘要：
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.
引用
收藏
页码：90 / +
页数：2
相关论文
共 50 条
  • [1] LINEAR-PROGRAMMING FORMULATIONS OF MARKOV DECISION-PROCESSES
    NAZARETH, JL
    KULKARNI, RB
    OPERATIONS RESEARCH LETTERS, 1986, 5 (01) : 13 - 16
  • [2] Using Linear Programming for Bayesian Exploration in Markov Decision Processes
    Castro, Pablo Samuel
    Precup, Doina
    20TH INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE, 2007, : 2436 - 2441
  • [3] FINITE LINEAR PROGRAMMING APPROXIMATIONS OF CONSTRAINED DISCOUNTED MARKOV DECISION PROCESSES
    Dufour, Francois
    Prieto-Rumeau, Tomas
    SIAM JOURNAL ON CONTROL AND OPTIMIZATION, 2013, 51 (02) : 1298 - 1324
  • [4] Approximate Linear Programming for Constrained Partially Observable Markov Decision Processes
    Poupart, Pascal
    Malhotra, Aarti
    Pei, Pei
    Kim, Kee-Eung
    Goh, Bongseok
    Bowling, Michael
    PROCEEDINGS OF THE TWENTY-NINTH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, 2015, : 3342 - 3348
  • [5] Quantitative Programming and Markov Decision Processes
    Todoran, Eneia Nicolae
    2022 24TH INTERNATIONAL SYMPOSIUM ON SYMBOLIC AND NUMERIC ALGORITHMS FOR SCIENTIFIC COMPUTING, SYNASC, 2022, : 117 - 124
  • [6] LINEAR PROGRAMMING IN A MARKOV DECISION PROBLEM
    DENARDO, EV
    MANAGEMENT SCIENCE SERIES A-THEORY, 1970, 16 (05): : 281 - 288
  • [7] Asymptotic linear programming and policy improvement for singularly perturbed Markov decision processes
    Eitan Altman
    Konstantin E. Avrachenkov
    Jerzy A. Filar
    Mathematical Methods of Operations Research, 1999, 49 (1) : 97 - 109
  • [8] Asymptotic linear programming and policy improvement for singularly perturbed Markov decision processes
    Altman, E
    Avrachenkov, KE
    Filar, JA
    MATHEMATICAL METHODS OF OPERATIONS RESEARCH, 1999, 49 (01) : 97 - 109
  • [9] The Linear Programming Approach to Reach-Avoid Problems for Markov Decision Processes
    Kariotoglou, Nikolaos
    Kamgarpour, Maryam
    Summers, Tyler H.
    Lygeros, John
    JOURNAL OF ARTIFICIAL INTELLIGENCE RESEARCH, 2017, 60 : 263 - 285
  • [10] Conditions for the Solvability of the Linear Programming Formulation for Constrained Discounted Markov Decision Processes
    F. Dufour
    T. Prieto-Rumeau
    Applied Mathematics & Optimization, 2016, 74 : 27 - 51
12345