Agent behavior monitoring using optimal action selection and twin gaussian processes

Autores
Avila, Luis; Martínez, Ernesto
Año de publicación
2014
Idioma
inglés
Tipo de recurso
documento de conferencia
Estado
versión publicada
Descripción
The increasing trend towards delegating complex tasks to autonomous artificial agents in safety-critical socio-technical systems makes agent behavior monitoring of paramount importance. In this work, a probabilistic approach for on-line monitoring using optimal action selection and twin Gaussian processes (TGP) is proposed. A Kullback-Leibler (KL) based metric is proposed to characterize the deviation of an agent behavior (modeled as a controlled stochastic process) to its specification. The optimal behavior specification is obtained using Linearly Solvable Markov Decision Processes (LSMDP) whereby the Bellman equation is made linear through an exponential transformation such that the optimal control policy is obtained in an explicit form.
Sociedad Argentina de Informática e Investigación Operativa (SADIO)
Materia
Ciencias Informáticas
agent monitoring
gaussian processes
optimal selection
Nivel de accesibilidad
acceso abierto
Condiciones de uso
http://creativecommons.org/licenses/by/3.0/
Repositorio
SEDICI (UNLP)
Institución
Universidad Nacional de La Plata
OAI Identificador
oai:sedici.unlp.edu.ar:10915/41659

id SEDICI_7a754a57d60be129cb87d56102fd2960
oai_identifier_str oai:sedici.unlp.edu.ar:10915/41659
network_acronym_str SEDICI
repository_id_str 1329
network_name_str SEDICI (UNLP)
spelling Agent behavior monitoring using optimal action selection and twin gaussian processesAvila, LuisMartínez, ErnestoCiencias Informáticasagent monitoringgaussian processesoptimal selectionThe increasing trend towards delegating complex tasks to autonomous artificial agents in safety-critical socio-technical systems makes agent behavior monitoring of paramount importance. In this work, a probabilistic approach for on-line monitoring using optimal action selection and twin Gaussian processes (TGP) is proposed. A Kullback-Leibler (KL) based metric is proposed to characterize the deviation of an agent behavior (modeled as a controlled stochastic process) to its specification. The optimal behavior specification is obtained using Linearly Solvable Markov Decision Processes (LSMDP) whereby the Bellman equation is made linear through an exponential transformation such that the optimal control policy is obtained in an explicit form.Sociedad Argentina de Informática e Investigación Operativa (SADIO)2014-11info:eu-repo/semantics/conferenceObjectinfo:eu-repo/semantics/publishedVersionObjeto de conferenciahttp://purl.org/coar/resource_type/c_5794info:ar-repo/semantics/documentoDeConferenciaapplication/pdf1-8http://sedici.unlp.edu.ar/handle/10915/41659enginfo:eu-repo/semantics/altIdentifier/url/http://43jaiio.sadio.org.ar/proceedings/ASAI/1.pdfinfo:eu-repo/semantics/altIdentifier/issn/1850-2784info:eu-repo/semantics/openAccesshttp://creativecommons.org/licenses/by/3.0/Creative Commons Attribution 3.0 Unported (CC BY 3.0)reponame:SEDICI (UNLP)instname:Universidad Nacional de La Platainstacron:UNLP2025-09-29T11:01:08Zoai:sedici.unlp.edu.ar:10915/41659Institucionalhttp://sedici.unlp.edu.ar/Universidad públicaNo correspondehttp://sedici.unlp.edu.ar/oai/snrdalira@sedici.unlp.edu.arArgentinaNo correspondeNo correspondeNo correspondeopendoar:13292025-09-29 11:01:08.664SEDICI (UNLP) - Universidad Nacional de La Platafalse
dc.title.none.fl_str_mv Agent behavior monitoring using optimal action selection and twin gaussian processes
title Agent behavior monitoring using optimal action selection and twin gaussian processes
spellingShingle Agent behavior monitoring using optimal action selection and twin gaussian processes
Avila, Luis
Ciencias Informáticas
agent monitoring
gaussian processes
optimal selection
title_short Agent behavior monitoring using optimal action selection and twin gaussian processes
title_full Agent behavior monitoring using optimal action selection and twin gaussian processes
title_fullStr Agent behavior monitoring using optimal action selection and twin gaussian processes
title_full_unstemmed Agent behavior monitoring using optimal action selection and twin gaussian processes
title_sort Agent behavior monitoring using optimal action selection and twin gaussian processes
dc.creator.none.fl_str_mv Avila, Luis
Martínez, Ernesto
author Avila, Luis
author_facet Avila, Luis
Martínez, Ernesto
author_role author
author2 Martínez, Ernesto
author2_role author
dc.subject.none.fl_str_mv Ciencias Informáticas
agent monitoring
gaussian processes
optimal selection
topic Ciencias Informáticas
agent monitoring
gaussian processes
optimal selection
dc.description.none.fl_txt_mv The increasing trend towards delegating complex tasks to autonomous artificial agents in safety-critical socio-technical systems makes agent behavior monitoring of paramount importance. In this work, a probabilistic approach for on-line monitoring using optimal action selection and twin Gaussian processes (TGP) is proposed. A Kullback-Leibler (KL) based metric is proposed to characterize the deviation of an agent behavior (modeled as a controlled stochastic process) to its specification. The optimal behavior specification is obtained using Linearly Solvable Markov Decision Processes (LSMDP) whereby the Bellman equation is made linear through an exponential transformation such that the optimal control policy is obtained in an explicit form.
Sociedad Argentina de Informática e Investigación Operativa (SADIO)
description The increasing trend towards delegating complex tasks to autonomous artificial agents in safety-critical socio-technical systems makes agent behavior monitoring of paramount importance. In this work, a probabilistic approach for on-line monitoring using optimal action selection and twin Gaussian processes (TGP) is proposed. A Kullback-Leibler (KL) based metric is proposed to characterize the deviation of an agent behavior (modeled as a controlled stochastic process) to its specification. The optimal behavior specification is obtained using Linearly Solvable Markov Decision Processes (LSMDP) whereby the Bellman equation is made linear through an exponential transformation such that the optimal control policy is obtained in an explicit form.
publishDate 2014
dc.date.none.fl_str_mv 2014-11
dc.type.none.fl_str_mv info:eu-repo/semantics/conferenceObject
info:eu-repo/semantics/publishedVersion
Objeto de conferencia
http://purl.org/coar/resource_type/c_5794
info:ar-repo/semantics/documentoDeConferencia
format conferenceObject
status_str publishedVersion
dc.identifier.none.fl_str_mv http://sedici.unlp.edu.ar/handle/10915/41659
url http://sedici.unlp.edu.ar/handle/10915/41659
dc.language.none.fl_str_mv eng
language eng
dc.relation.none.fl_str_mv info:eu-repo/semantics/altIdentifier/url/http://43jaiio.sadio.org.ar/proceedings/ASAI/1.pdf
info:eu-repo/semantics/altIdentifier/issn/1850-2784
dc.rights.none.fl_str_mv info:eu-repo/semantics/openAccess
http://creativecommons.org/licenses/by/3.0/
Creative Commons Attribution 3.0 Unported (CC BY 3.0)
eu_rights_str_mv openAccess
rights_invalid_str_mv http://creativecommons.org/licenses/by/3.0/
Creative Commons Attribution 3.0 Unported (CC BY 3.0)
dc.format.none.fl_str_mv application/pdf
1-8
dc.source.none.fl_str_mv reponame:SEDICI (UNLP)
instname:Universidad Nacional de La Plata
instacron:UNLP
reponame_str SEDICI (UNLP)
collection SEDICI (UNLP)
instname_str Universidad Nacional de La Plata
instacron_str UNLP
institution UNLP
repository.name.fl_str_mv SEDICI (UNLP) - Universidad Nacional de La Plata
repository.mail.fl_str_mv alira@sedici.unlp.edu.ar
_version_ 1844615877463375872
score 13.070432