Publication:
A unified framework for linear function approximation of value functions in stochastic control

Simple item page
dc.affiliation.dptoUC3M. Departamento de Teoría de la Señal y Comunicacioneses
dc.affiliation.grupoinvUC3M. Grupo de Investigación: Comunicacioneses
dc.contributor.authorSánchez-Fernández, Matilde
dc.contributor.authorValcárcel, Sergio
dc.contributor.authorZazo, Santiago
dc.date.accessioned2015-06-26T11:52:21Z
dc.date.available2015-06-26T11:52:21Z
dc.date.issued2013-09
dc.descriptionThe proceeding at:21st European Signal Processing Conference (EUSIPCO 2013), took place 2013, September 9-13, in Marrakech (marroc).en
dc.description.abstractThis paper contributes with a unified formulation that merges previous 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 approximated, 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 proposed algorithm is illustrated by simulation, showing competitive results when compared with the state-of-the-art solutions.en
dc.description.sponsorshipThis work has been partly funded by the Spanish Ministry of Science and Innovation with the project GRE3N (TEC 2011-29006-C03-01/02/03) and in the program CONSOLIDER-INGENIO 2010 under project COMONSENS (CSD 2008-00010). This work was supported in part by the Spanish Ministry of Science and Innovation under the grants TEC2009-14219-C03-01,TEC2010-21217-C02- 02-CR4HFDVL and in the program CONSOLIDER-INGENIO 2010 under the grant CSD2008-00010 COMONSENS; and by the European Commission under the grant FP7-ICT-2009-4-248894-WHERE-2.en
dc.description.statusPublicado
dc.format.extent5
dc.format.mimetypeapplication/pdf
dc.identifier.bibliographicCitationProceedings of the 21st European Signal Processing Conference (EUSIPCO) (2013) pp. 1-5en
dc.identifier.publicationfirstpage1
dc.identifier.publicationlastpage5
dc.identifier.publicationtitleProceedings of the 21st European Signal Processing Conference (EUSIPCO)en
dc.identifier.urihttps://hdl.handle.net/10016/21205
dc.identifier.uxxiCC/0000022069
dc.language.isoeng
dc.publisherIEEE - The Institute of Electrical and Electronics Engineers, Incen
dc.relation.eventdate2013, September 9-13en
dc.relation.eventnumber21
dc.relation.eventplaceMarrakech (Marroc)en
dc.relation.eventtitleEuropean Signal Processing Conference (EUSIPCO 2013)en
dc.relation.projectIDGobierno de España. TEC2011-29006-C03-02es
dc.relation.projectIDGobierno de España. TEC2011-29006-C03-03es
dc.relation.publisherversionhttp://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=6811729
dc.rights© 2013 IEEEen
dc.rights.accessRightsopen access
dc.subject.ecienciaTelecomunicacioneses
dc.subject.otherFunction approximationen
dc.subject.otherMarkov processesen
dc.subject.otherSignal processingen
dc.subject.otherStochastic systemsen
dc.titleA unified framework for linear function approximation of value functions in stochastic controlen
dc.typeconference paper*
dc.type.hasVersionAM*
dspace.entity.typePublication
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
unied_EUSIPCO_2013_ps.pdf
Size:
718.99 KB
Format:
Adobe Portable Document Format
Download
Collections
DTSC - GC - Comunicaciones en congresos y otros eventos
DTSC - GC - Capítulos de Monografías