Texto completo
![[thumbnail of INVE_MEM_2013_166531.pdf]](https://oa.upm.es/style/images/fileicons/application_pdf.png) Vista Previa |
PDF (Portable Document Format)
- Se necesita un visor de ficheros PDF, como GSview, Xpdf o Adobe Acrobat Reader
Descargar (1MB) |
A unified framework for linear function approximation of value functions in stochastic control
Cita
Sánchez Fernández, Matilde, Valcarcel Macua, Sergio and Zazo Bello, Santiago 
ORCID: https://orcid.org/0000-0001-9073-7927
(2013).
A unified framework for linear function approximation of value functions in stochastic control.
En: "21st European Signal Processing Conference (EUSIPCO)", 09/09/2013 - 13/09/2013, Marrakech, Morocco. pp. 1-5.
ORCID: https://orcid.org/0000-0001-9073-7927
(2013).
A unified framework for linear function approximation of value functions in stochastic control.
En: "21st European Signal Processing Conference (EUSIPCO)", 09/09/2013 - 13/09/2013, Marrakech, Morocco. pp. 1-5.
Descripción
| Título: | A unified framework for linear function approximation of value functions in stochastic control |
|---|---|
| Autor/es: |
|
| Tipo de Documento: | Ponencia en Congreso o Jornada (Artículo) |
| Título del Evento: | 21st European Signal Processing Conference (EUSIPCO) |
| Fechas del Evento: | 09/09/2013 - 13/09/2013 |
| Lugar del Evento: | Marrakech, Morocco |
| Título del Libro: | 21st European Signal Processing Conference (EUSIPCO) |
| Fecha: | 2013 |
| Materias: | |
| ODS: | |
| Palabras Clave Informales: | Approximate dynamic programming, Linear value function approximation, Mean squared Bellman Error, Mean squared projected Bellman Error, Reinforcement Learning |
| Escuela: | E.T.S.I. Telecomunicación (UPM) |
| Departamento: | Señales, Sistemas y Radiocomunicaciones |
| Licencias Creative Commons: | Reconocimiento - Sin obra derivada - No comercial |
Resumen
This paper contributes with a unified formulation that merges previ- ous analysis on the prediction of the performance ( value function ) of certain sequence of actions ( policy ) when an agent operates a Markov decision process with large state-space. When the states are represented by features and the value function is linearly approxi- mated, our analysis reveals a new relationship between two common cost functions used to obtain the optimal approximation. In addition, this analysis allows us to propose an efficient adaptive algorithm that provides an unbiased linear estimate. The performance of the pro- posed algorithm is illustrated by simulation, showing competitive results when compared with the state-of-the-art solutions.
Más información
| ID de Registro: | 28942 |
|---|---|
| Identificador DC: | https://oa.upm.es/28942/ |
| Identificador OAI: | oai:oa.upm.es:28942 |
| URL Oficial: | http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumb... |
| Depositado por: | Memoria Investigacion |
| Depositado el: | 30 Jun 2014 16:04 |
| Ultima Modificación: | 22 Sep 2014 11:43 |
Acciones
- Estadísticas
- Exportar cita
- Editar (sólo personal del Archivo)
