The linear programming approach in constrained Markov decision processes 条件付きマルコフ決定過程における線形計画法

Access this Article

Title: The linear programming approach in constrained Markov decision processes

博士論文

Contents / p1 (0004.jp2)
Preface / p3 (0006.jp2)
Summary of Notation and Abbreviations / p5 (0008.jp2)
1 Introduction / p1 (0011.jp2)
1.1 Constrained MDPs Model / p1 (0011.jp2)
1.2 Policies and Performance Criteria / p4 (0014.jp2)
1.3 Optimization Problem / p7 (0017.jp2)
1.4 Linear Programs / p8 (0018.jp2)
1.5 Duality Gap / p12 (0022.jp2)
2 Long-Run Average with Bounded Reward / p14 (0024.jp2)
2.1 Introduction / p15 (0025.jp2)
2.2 Solvability and Absence of Duality Gap / p18 (0028.jp2)
2.3 Existence of Optimal Stationary Policy / p25 (0035.jp2)
3 Long-Run Average with Unbounded Reward / p30 (0040.jp2)
3.1 Introduction / p30 (0040.jp2)
3.2 Solvability and Absence of Duality Gap / p33 (0043.jp2)
3.3 Existence of Optimal Stationary Policy / p39 (0049.jp2)
4 Sample-Path Average with General Space / p44 (0054.jp2)
4.1 Introduction / p45 (0055.jp2)
4.2 Optimal Pair and Linear Program / p46 (0056.jp2)
4.3 State-Wise Mixed Stationary Policies / p49 (0059.jp2)
Appendix / p55 (0065.jp2)
Acknowlegements / p59 (0069.jp2)
Bibliography / p60 (0070.jp2)

2access