| bibtype |
C -
Conference Paper (international conference)
|
| ARLID |
0333145 |
| utime |
20240111140729.9 |
| mtime |
20091210235959.9 |
| title
(primary) (eng) |
Solving Growth Rates and Average Optimality in Risk-Sensitive Markov Decision Chains |
| specification |
| page_count |
34 s. |
| media_type |
www |
|
| serial |
| ARLID |
cav_un_epca*0333215 |
| title
|
Proceedings of 1st International Conference on Applied Mathematics |
| page_num |
27-60 |
| publisher |
| place |
Almoloya de Juárey, Edo de México |
| name |
Universidad Politécnica del Valle de Toluca |
| year |
2009 |
|
| editor |
| name1 |
Díaz Nunez |
| name2 |
J. J. |
|
| editor |
| name1 |
Carmona |
| name2 |
R. G. |
|
| editor |
|
| editor |
|
| editor |
| name1 |
Rodríguez |
| name2 |
M. A. |
|
| editor |
| name1 |
ére González |
| name2 |
L. A. |
|
| editor |
| name1 |
Castillo Rubi |
| name2 |
M. A. |
|
|
| title
(cze) |
Výpočet míry růstu and optimality v průměru v markovských rozhodovacích řetězcích za rizika |
| keyword |
risk-sensitive Markov decision chains |
| keyword |
growth rate |
| keyword |
average optimality |
| author
(primary) |
| ARLID |
cav_un_auth*0101196 |
| name1 |
Sladký |
| name2 |
Karel |
| institution |
UTIA-B |
| full_dept |
Department of Econometrics |
| fullinstit |
Ústav teorie informace a automatizace AV ČR, v. v. i. |
|
| source |
|
| cas_special |
| project |
| project_id |
GA402/08/0107 |
| agency |
GA ČR |
| country |
CZ |
| ARLID |
cav_un_auth*0240545 |
|
| research |
CEZ:AV0Z10750506 |
| abstract
(eng) |
The study of risk-sensitive Markov decision processes is usually restricted to processes with a single class of recurrent state and no transient states. Moreover, the analysis was restricted only to discrete-time models, only little attention, if any, was devoted to the continuous-time models. The aim of this article is to extend the analysis to models with transient states as well as to multichain Markov processes both in discrete- and continuous-time setting. Our analysis is based on more general models of stochastic dynamic programming where transition probability matrices are replaced by general nonnegative matrices (discrete-time case) or by matrices with nonnegative off-diagonal entries (continuous-time case). |
| abstract
(cze) |
Problematika markovských rozhodovacích procesů za přítomnosti rizika se obvykle studuje pouze pro nerozložitelné procesy a v existující literatuře byly pouze studovány procesy s diskrétním časovým parametrem. Cílem této práce je rozšířít analýzu těchto procesů na modely s tranzientními stavy a dále na obecný případ markovských procesů jak pro diskrétní, tak i pro spojitý časový parametr. Použitá metodika řešení je založena na analýze zobecněných modelů stochastického dynamického programování, kde pravděpodobnosti přechodu jsou nahrazeny obecnými nezápornými maticemi (modely s diskrétní časovým parametrem) nebo maticemi s nezápornými elementy mimo hlavní diagonálu (modely ve spojitém čase). |
| action |
| ARLID |
cav_un_auth*0256674 |
| name |
1st International Conference on Applied Mathematics |
| place |
Almoloya, Edo. de México |
| dates |
04.11.2009-06.11.2009 |
| country |
MX |
|
| reportyear |
2010 |
| RIV |
BB |
| permalink |
http://hdl.handle.net/11104/0178208 |
| arlyear |
2009 |
| mrcbU56 |
pdf |
| mrcbU63 |
cav_un_epca*0333215 Proceedings of 1st International Conference on Applied Mathematics 27 60 Almoloya de Juárey, Edo de México Universidad Politécnica del Valle de Toluca 2009 |
| mrcbU67 |
Díaz Nunez J. J. 340 |
| mrcbU67 |
Carmona R. G. 340 |
| mrcbU67 |
Leal J. S. 340 |
| mrcbU67 |
Ramos C. D. 340 |
| mrcbU67 |
Rodríguez M. A. 340 |
| mrcbU67 |
ére González L. A. 340 |
| mrcbU67 |
Castillo Rubi M. A. 340 |
|