1. Repository of OAI
  2. ΑΠΟΘΕΤΗΡΙΟ "ΟΛΥΜΠΙΑΣ"
  3. Σχολή Θετικών Επιστημών
  4. Τμήμα Μηχανικών Ηλεκτρονικών Υπολογιστών και Πληροφορικής
  5. Άρθρα σε επιστημονικά περιοδικά ( Ανοικτά)
Ελληνικά   English  
Please use this identifier to cite or link to this item: https://olympias.lib.uoi.gr/jspui/handle/123456789/11059
Full metadata record
DC FieldValueLanguage
dc.contributor.authorBlekas, K.en
dc.contributor.authorTziortziotis, N.en
dc.date.accessioned2015-11-24T17:02:29Z-
dc.date.available2015-11-24T17:02:29Z-
dc.identifier.urihttps://olympias.lib.uoi.gr/jspui/handle/123456789/11059-
dc.rightsDefault Licence-
dc.titleA bayesian reinforcement learning framework using relevant vector machinesen
heal.typejournalArticle-
heal.type.enJournal articleen
heal.type.elΆρθρο Περιοδικούel
heal.languageen-
heal.accesscampus-
heal.recordProviderΠανεπιστήμιο Ιωαννίνων. Σχολή Θετικών Επιστημών. Τμήμα Μηχανικών Ηλεκτρονικών Υπολογιστών και Πληροφορικήςel
heal.publicationDate2011-
heal.abstractIn this work we present an advanced Bayesian formulation to the task of control learning that employs the Relevance Vector Machines (RVM) generative model for value function evaluation. The key aspect of the proposed method is the design of the discount return as a generalized linear model that constitutes a wellknown probabilistic approach. This allows to augment the model with advantageous sparse priors provided by the RVM s regression framework. We have also taken into account the significant issue of selecting the proper parameters of the kernel design matrix. Experiments have shown that our method produces improved performance in both simulated and real test environments.en
heal.journalNameAAAIen
heal.journalTypepeer reviewed-
heal.fullTextAvailabilityTRUE-
Appears in Collections:Άρθρα σε επιστημονικά περιοδικά ( Ανοικτά)

Show simple item record
Files in This Item:
File Description SizeFormat 
blekas-2011-A Bayesian Reinforcement Learning.pdf454.02 kBAdobe PDFView/Open    Request a copy
Show simple item record  


This item is licensed under a Creative Commons License Creative Commons