Parameter Estimation Problems in Markov Random Processes

Результат исследований: Публикации в книгах, отчётах, сборниках, трудах конференцийстатья в сборнике материалов конференциинаучнаярецензирование

Выдержка

Problems of convergence and stability of Bayesian estimates in the identification of stochastic control systems are considered. The informational measure of the mismatch between the estimated distribution and the estimate is the main apparatus for establishing the fact of convergence. The choice of a priori distribution of parameters is not always obvious. The Kullback-Leibler information number is taken as such measure. The convergence of the estimates of the transition function of the process to the non-stationary transition function is established in this paper. The problem of synthesis of optimal strategies for dynamic systems in which there is no part of the main information needed for constructing the optimal control is also considered. It is assumed that the system contains at least one unknown parameter belonging to some parameter space. Therefore, the class of control systems considered in the article is the class of parametric adaptive systems.

Язык оригиналаанглийский
Название основной публикацииComputational Science and Its Applications- ICCSA 2019 - 19th International Conference, Proceedings
РедакторыBeniamino Murgante, Osvaldo Gervasi, Elena Stankova, Vladimir Korkhov, Sanjay Misra, Carmelo Torre, Eufemia Tarantino, David Taniar, Ana Maria A.C. Rocha, Bernady O. Apduhan
ИздательSpringer
Страницы691-701
Число страниц11
ISBN (печатное издание)9783030243043
DOI
СостояниеОпубликовано - 1 янв 2019
Событие19th International Conference on Computational Science and Its Applications, ICCSA 2019 - Saint Petersburg, Российская Федерация
Продолжительность: 1 июл 20194 июл 2019

Серия публикаций

НазваниеLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Том11622 LNCS
ISSN (печатное издание)0302-9743
ISSN (электронное издание)1611-3349

Конференция

Конференция19th International Conference on Computational Science and Its Applications, ICCSA 2019
СтранаРоссийская Федерация
ГородSaint Petersburg
Период1/07/194/07/19

Отпечаток

Random process
Random processes
Markov Process
Parameter estimation
Parameter Estimation
Stochastic control systems
Adaptive systems
Control System
Estimate
Kullback-Leibler Information
Identification (control systems)
Dynamical systems
Stochastic Control
Stability and Convergence
Adaptive Systems
Optimal Strategy
Control systems
Stochastic Systems
Unknown Parameters
Dynamic Systems

Предметные области Scopus

  • Теоретические компьютерные науки
  • Компьютерные науки (все)

Цитировать

Karelin, V., Fominyh, A., Myshkov, S., & Polyakova, L. (2019). Parameter Estimation Problems in Markov Random Processes. В B. Murgante, O. Gervasi, E. Stankova, V. Korkhov, S. Misra, C. Torre, E. Tarantino, D. Taniar, A. M. A. C. Rocha, ... B. O. Apduhan (Ред.), Computational Science and Its Applications- ICCSA 2019 - 19th International Conference, Proceedings (стр. 691-701). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Том 11622 LNCS). Springer. https://doi.org/10.1007/978-3-030-24305-0_51
Karelin, Vladimir ; Fominyh, Alexander ; Myshkov, Stanislav ; Polyakova, Lyudmila. / Parameter Estimation Problems in Markov Random Processes. Computational Science and Its Applications- ICCSA 2019 - 19th International Conference, Proceedings. редактор / Beniamino Murgante ; Osvaldo Gervasi ; Elena Stankova ; Vladimir Korkhov ; Sanjay Misra ; Carmelo Torre ; Eufemia Tarantino ; David Taniar ; Ana Maria A.C. Rocha ; Bernady O. Apduhan. Springer, 2019. стр. 691-701 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{5b33faa8ce754ef6a8ecfb0f322bcd5c,
title = "Parameter Estimation Problems in Markov Random Processes",
abstract = "Problems of convergence and stability of Bayesian estimates in the identification of stochastic control systems are considered. The informational measure of the mismatch between the estimated distribution and the estimate is the main apparatus for establishing the fact of convergence. The choice of a priori distribution of parameters is not always obvious. The Kullback-Leibler information number is taken as such measure. The convergence of the estimates of the transition function of the process to the non-stationary transition function is established in this paper. The problem of synthesis of optimal strategies for dynamic systems in which there is no part of the main information needed for constructing the optimal control is also considered. It is assumed that the system contains at least one unknown parameter belonging to some parameter space. Therefore, the class of control systems considered in the article is the class of parametric adaptive systems.",
keywords = "Bayesian probability theory, Kullback-Leibler information number, Parameter estimation",
author = "Vladimir Karelin and Alexander Fominyh and Stanislav Myshkov and Lyudmila Polyakova",
year = "2019",
month = "1",
day = "1",
doi = "10.1007/978-3-030-24305-0_51",
language = "English",
isbn = "9783030243043",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer",
pages = "691--701",
editor = "Beniamino Murgante and Osvaldo Gervasi and Elena Stankova and Vladimir Korkhov and Sanjay Misra and Carmelo Torre and Eufemia Tarantino and David Taniar and Rocha, {Ana Maria A.C.} and Apduhan, {Bernady O.}",
booktitle = "Computational Science and Its Applications- ICCSA 2019 - 19th International Conference, Proceedings",
address = "Germany",

}

Karelin, V, Fominyh, A, Myshkov, S & Polyakova, L 2019, Parameter Estimation Problems in Markov Random Processes. в B Murgante, O Gervasi, E Stankova, V Korkhov, S Misra, C Torre, E Tarantino, D Taniar, AMAC Rocha & BO Apduhan (ред.), Computational Science and Its Applications- ICCSA 2019 - 19th International Conference, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), том. 11622 LNCS, Springer, стр. 691-701, Saint Petersburg, Российская Федерация, 1/07/19. https://doi.org/10.1007/978-3-030-24305-0_51

Parameter Estimation Problems in Markov Random Processes. / Karelin, Vladimir; Fominyh, Alexander; Myshkov, Stanislav; Polyakova, Lyudmila.

Computational Science and Its Applications- ICCSA 2019 - 19th International Conference, Proceedings. ред. / Beniamino Murgante; Osvaldo Gervasi; Elena Stankova; Vladimir Korkhov; Sanjay Misra; Carmelo Torre; Eufemia Tarantino; David Taniar; Ana Maria A.C. Rocha; Bernady O. Apduhan. Springer, 2019. стр. 691-701 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Том 11622 LNCS).

Результат исследований: Публикации в книгах, отчётах, сборниках, трудах конференцийстатья в сборнике материалов конференциинаучнаярецензирование

TY - GEN

T1 - Parameter Estimation Problems in Markov Random Processes

AU - Karelin, Vladimir

AU - Fominyh, Alexander

AU - Myshkov, Stanislav

AU - Polyakova, Lyudmila

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Problems of convergence and stability of Bayesian estimates in the identification of stochastic control systems are considered. The informational measure of the mismatch between the estimated distribution and the estimate is the main apparatus for establishing the fact of convergence. The choice of a priori distribution of parameters is not always obvious. The Kullback-Leibler information number is taken as such measure. The convergence of the estimates of the transition function of the process to the non-stationary transition function is established in this paper. The problem of synthesis of optimal strategies for dynamic systems in which there is no part of the main information needed for constructing the optimal control is also considered. It is assumed that the system contains at least one unknown parameter belonging to some parameter space. Therefore, the class of control systems considered in the article is the class of parametric adaptive systems.

AB - Problems of convergence and stability of Bayesian estimates in the identification of stochastic control systems are considered. The informational measure of the mismatch between the estimated distribution and the estimate is the main apparatus for establishing the fact of convergence. The choice of a priori distribution of parameters is not always obvious. The Kullback-Leibler information number is taken as such measure. The convergence of the estimates of the transition function of the process to the non-stationary transition function is established in this paper. The problem of synthesis of optimal strategies for dynamic systems in which there is no part of the main information needed for constructing the optimal control is also considered. It is assumed that the system contains at least one unknown parameter belonging to some parameter space. Therefore, the class of control systems considered in the article is the class of parametric adaptive systems.

KW - Bayesian probability theory

KW - Kullback-Leibler information number

KW - Parameter estimation

UR - http://www.scopus.com/inward/record.url?scp=85068591717&partnerID=8YFLogxK

UR - http://www.mendeley.com/research/parameter-estimation-problems-markov-random-processes

U2 - 10.1007/978-3-030-24305-0_51

DO - 10.1007/978-3-030-24305-0_51

M3 - Conference contribution

AN - SCOPUS:85068591717

SN - 9783030243043

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 691

EP - 701

BT - Computational Science and Its Applications- ICCSA 2019 - 19th International Conference, Proceedings

A2 - Murgante, Beniamino

A2 - Gervasi, Osvaldo

A2 - Stankova, Elena

A2 - Korkhov, Vladimir

A2 - Misra, Sanjay

A2 - Torre, Carmelo

A2 - Tarantino, Eufemia

A2 - Taniar, David

A2 - Rocha, Ana Maria A.C.

A2 - Apduhan, Bernady O.

PB - Springer

ER -

Karelin V, Fominyh A, Myshkov S, Polyakova L. Parameter Estimation Problems in Markov Random Processes. В Murgante B, Gervasi O, Stankova E, Korkhov V, Misra S, Torre C, Tarantino E, Taniar D, Rocha AMAC, Apduhan BO, редакторы, Computational Science and Its Applications- ICCSA 2019 - 19th International Conference, Proceedings. Springer. 2019. стр. 691-701. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-030-24305-0_51