Mostrar el registro sencillo del ítem
dc.contributor.author
Barsce, Juan Cruz
![Se ha confirmado la validez de este valor de autoridad por un usuario](/themes/CONICETDigital/images/authority_control/invisible.gif)
dc.contributor.author
Palombarini, Jorge Andrés
![Se ha confirmado la validez de este valor de autoridad por un usuario](/themes/CONICETDigital/images/authority_control/invisible.gif)
dc.contributor.author
Martínez, Ernesto Carlos
![Se ha confirmado la validez de este valor de autoridad por un usuario](/themes/CONICETDigital/images/authority_control/invisible.gif)
dc.date.available
2019-10-22T18:05:22Z
dc.date.issued
2018
dc.identifier.citation
Barsce, Juan Cruz; Palombarini, Jorge Andrés; Martínez, Ernesto Carlos; Towards Autonomous Reinforcement Learning: Automatic Setting of Hyper-parameters using Bayesian Optimization; CLEI (Latin-american Center for Informatics Studies); CLEI Electronic Journal; 21; 2; 2018; 1-22
dc.identifier.issn
0717-5000
dc.identifier.uri
http://hdl.handle.net/11336/86940
dc.description.abstract
With the increase of machine learning usage by industries and scientific communities in a variety of tasks such as text mining, image recognition and self-driving cars, automatic setting of hyper-parameter in learning algorithms is a key factor for obtaining good performances regardless of user expertise in the inner workings of the techniques and methodologies. In particular, for a reinforcement learning algorithm, the efficiency of an agent learning a control policy in an uncertain environment is heavily dependent on the hyper-parameters used to balance exploration with exploitation. In this work, an autonomous learning framework that integrates Bayesian optimization with Gaussian process regression to optimize the hyper-parameters of a reinforcement learning algorithm, is proposed. Also, a bandits-based approach to achieve a balance between computational costs and decreasing uncertainty about the extit{Q}-values, is presented. A gridworld example is used to highlight how hyper-parameter configurations of a learning algorithm (SARSA) are iteratively improved based on two performance functions.
dc.format
application/pdf
dc.language.iso
eng
dc.publisher
CLEI (Latin-american Center for Informatics Studies)
dc.rights
info:eu-repo/semantics/openAccess
dc.rights.uri
https://creativecommons.org/licenses/by-nc-sa/2.5/ar/
dc.subject
REINFORCEMENT LEARNING
dc.subject
AUTONOMOUS SYSTEMS
dc.subject
BAYESIAN OPTIMIZATION
dc.subject
HYPER-PARAMETERS SETTING
dc.subject.classification
Ingeniería de Sistemas y Comunicaciones
![Se ha confirmado la validez de este valor de autoridad por un usuario](/themes/CONICETDigital/images/authority_control/invisible.gif)
dc.subject.classification
Ingeniería Eléctrica, Ingeniería Electrónica e Ingeniería de la Información
![Se ha confirmado la validez de este valor de autoridad por un usuario](/themes/CONICETDigital/images/authority_control/invisible.gif)
dc.subject.classification
INGENIERÍAS Y TECNOLOGÍAS
![Se ha confirmado la validez de este valor de autoridad por un usuario](/themes/CONICETDigital/images/authority_control/invisible.gif)
dc.title
Towards Autonomous Reinforcement Learning: Automatic Setting of Hyper-parameters using Bayesian Optimization
dc.type
info:eu-repo/semantics/article
dc.type
info:ar-repo/semantics/artículo
dc.type
info:eu-repo/semantics/publishedVersion
dc.date.updated
2019-10-22T17:39:31Z
dc.journal.volume
21
dc.journal.number
2
dc.journal.pagination
1-22
dc.journal.pais
Uruguay
![Se ha confirmado la validez de este valor de autoridad por un usuario](/themes/CONICETDigital/images/authority_control/invisible.gif)
dc.journal.ciudad
Montevideo
dc.description.fil
Fil: Barsce, Juan Cruz. Universidad Tecnológica Nacional; Argentina
dc.description.fil
Fil: Palombarini, Jorge Andrés. Universidad Tecnológica Nacional; Argentina
dc.description.fil
Fil: Martínez, Ernesto Carlos. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - Santa Fe. Instituto de Desarrollo y Diseño. Universidad Tecnológica Nacional. Facultad Regional Santa Fe. Instituto de Desarrollo y Diseño; Argentina
dc.journal.title
CLEI Electronic Journal
dc.relation.alternativeid
info:eu-repo/semantics/altIdentifier/url/http://www.clei.org/cleiej-beta/index.php/cleiej/article/view/33
dc.relation.alternativeid
info:eu-repo/semantics/altIdentifier/doi/http://dx.doi.org/10.19153/cleiej.21.2.1
Archivos asociados