Read more
This book covers formulation, algorithms, and structural results of partially observed Markov decision processes, linking theory to real-world applications in controlled sensing.
List of contents
Preface; 1. Introduction; Part I. Stochastic Models and Bayesian Filtering: 2. Stochastic state-space models; 3. Optimal filtering; 4. Algorithms for maximum likelihood parameter estimation; 5. Multi-agent sensing: social learning and data incest; Part II. Partially Observed Markov Decision Processes. Models and Algorithms: 6. Fully observed Markov decision processes; 7. Partially observed Markov decision processes (POMDPs); 8. POMDPs in controlled sensing and sensor scheduling; Part III. Partially Observed Markov Decision Processes: 9. Structural results for Markov decision processes; 10. Structural results for optimal filters; 11. Monotonicity of value function for POMPDs; 12. Structural results for stopping time POMPDs; 13. Stopping time POMPDs for quickest change detection; 14. Myopic policy bounds for POMPDs and sensitivity to model parameters; Part IV. Stochastic Approximation and Reinforcement Learning: 15. Stochastic optimization and gradient estimation; 16. Reinforcement learning; 17. Stochastic approximation algorithms: examples; 18. Summary of algorithms for solving POMPDs; Appendix A. Short primer on stochastic simulation; Appendix B. Continuous-time HMM filters; Appendix C. Markov processes; Appendix D. Some limit theorems; Bibliography; Index.
About the author
Vikram Krishnamurthy is a Professor and Canada Research Chair in Statistical Signal Processing at the University of British Columbia, Vancouver. His research contributions focus on nonlinear filtering, stochastic approximation algorithms and POMDPs. Dr Krishnamurthy is a Fellow of the Institute of Electrical and Electronics Engineers (IEEE) and served as a distinguished lecturer for the IEEE Signal Processing Society. In 2013, he received an honorary doctorate from KTH, Royal Institute of Technology, Sweden.
Summary
This book covers formulation, algorithms, and structural results of partially observed Markov decision processes, whilst linking theory to real-world applications in controlled sensing. Computations are kept to a minimum, enabling students and researchers in engineering, operations research, and economics to understand the methods and determine the structure of their optimal solution.
