Towards Autonomous Reinforcement Learning: Automatic Setting of Hyper-parameters using Bayesian Optimization | ||
| dc.conicet.organizacion | Universidad Tecnológica Nacional | |
| dc.conicet.organizacion | Universidad Tecnológica Nacional | |
| dc.conicet.organizacion | 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 | |
| dc.contributor.author | Barsce, Juan Cruz | |
| dc.contributor.author | Palombarini, Jorge Andrés | |
| dc.contributor.author | Martínez, Ernesto Carlos | |
| dc.date.issued | 2018 | |
| 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. | es |
| 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.format | application/pdf | es |
| 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 | es |
| dc.identifier.institucional | CONICET | |
| dc.identifier.issn | 0717-5000 | es |
| dc.identifier.repo | CONICET Digital | |
| dc.identifier.uri | http://hdl.handle.net/11336/86940 | |
| dc.journal.ciudad | Montevideo | es |
| dc.journal.number | 2 | es |
| dc.journal.pagination | 1-22 | es |
| dc.journal.pais | Uruguay | es |
| dc.journal.title | CLEI Electronic Journal | es |
| dc.journal.volume | 21 | es |
| dc.language.iso | eng | es |
| dc.publisher | CLEI (Latin-american Center for Informatics Studies) | es |
| 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 | |
| dc.rights | info:eu-repo/semantics/openAccess | * |
| dc.rights.uri | https://creativecommons.org/licenses/by-nc-sa/2.5/ar/ | * |
| dc.subject | REINFORCEMENT LEARNING | es |
| dc.subject | AUTONOMOUS SYSTEMS | es |
| dc.subject | BAYESIAN OPTIMIZATION | es |
| dc.subject | HYPER-PARAMETERS SETTING | es |
| dc.subject.classification | Ingeniería de Sistemas y Comunicaciones | es |
| dc.subject.classification | Ingeniería Eléctrica, Ingeniería Electrónica e Ingeniería de la Información | es |
| dc.subject.classification | INGENIERÍAS Y TECNOLOGÍAS | es |
| dc.title | Towards Autonomous Reinforcement Learning: Automatic Setting of Hyper-parameters using Bayesian Optimization | es |
| dc.type | info:eu-repo/semantics/article | |
| dc.type | info:ar-repo/semantics/artículo | |
| dc.type | info:eu-repo/semantics/publishedVersion | |
Archivos
Bloque original
Cargando...
- Nombre:
- CONICET_Digital_Nro.9e5398bb-5b73-4c7b-9cdc-6b892fc0a819_A.pdf
- Tamaño:
- 718.65 KB
- Formato:
- Adobe Portable Document Format