Please use this identifier to cite or link to this item: https://hdl.handle.net/11147/3467
Full metadata record
DC FieldValueLanguage
dc.contributor.advisorÖzdemir, Serhanen
dc.contributor.authorKarakurt, Murat-
dc.date.accessioned2014-07-22T13:51:35Z-
dc.date.available2014-07-22T13:51:35Z-
dc.date.issued2003en
dc.identifier.urihttp://hdl.handle.net/11147/3467-
dc.descriptionThesis (Master)--Izmir Institute of Technology, Mechanical Engineering, Izmir, 2003en
dc.descriptionIncludes bibliographical references (leaves: 61-66)en
dc.descriptionText in English; Abstract: Turkish and Englishen
dc.descriptionvi, 75 leavesen
dc.description.abstractThis research has aspired to build a system which is capable of solving problems by means of its past experience, especially an autonomous agent that can learn from trial and error sequences. To achieve this, connectionist neural network architectures are combined with the reinforcement learning methods. And the credit assignment problem in multi layer perceptron (MLP) architectures is altered. In classical credit assignment problems, actual output of the system and the previously known data in which the system tries to approximate are compared and the discrepancy between them is attempted to be minimized. However, temporal difference credit assignment depends on the temporary successive outputs. By this new method, it is more feasible to find the relation between each event rather than their consequences.Also in this thesis k-means algorithm is modified. Moreover MLP architectures is written in C++ environment, like Backpropagation, Radial Basis Function Networks, Radial Basis Function Link Net, Self-organized neural network, k-means algorithm.And with their combination for the Reinforcement learning, temporal difference learning, and Q-learning architectures were realized, all these algorithms are simulated, and these simulations are created in C++ environment.As a result, reinforcement learning methods used have two main disadvantages during the process of creating autonomous agent. Firstly its training time is too long, and too many input parameters are needed to train the system. Hence it is seen that hardware implementation is not feasible yet. Further research is considered necessary.en
dc.language.isoenen_US
dc.publisherIzmir Institute of Technologyen
dc.rightsinfo:eu-repo/semantics/openAccessen_US
dc.subject.lccTJ223.P4 K37 2003en
dc.subject.lcshPerceptronsen
dc.subject.lcshReinforcement learning (Machine learning)en
dc.subject.lcshNeural networks (Computer science)en
dc.titleData driven modeling using reinforcement learning in autonomous agentsen_US
dc.typeMaster Thesisen_US
dc.institutionauthorKarakurt, Murat-
dc.departmentIzmir Institute of Technology. Mechanical Engineeringen
dc.relation.publicationcategoryTezen_US
item.cerifentitytypePublications-
item.openairetypeMaster Thesis-
item.languageiso639-1en-
item.openairecristypehttp://purl.org/coar/resource_type/c_18cf-
item.grantfulltextopen-
item.fulltextWith Fulltext-
Appears in Collections:Master Degree / Yüksek Lisans Tezleri
Sürdürülebilir Yeşil Kampüs Koleksiyonu / Sustainable Green Campus Collection
Files in This Item:
File Description SizeFormat 
T000276.pdfMasterThesis1.12 MBAdobe PDFThumbnail
View/Open
Show simple item record

CORE Recommender

Page view(s)

80
checked on Jan 30, 2023

Download(s)

24
checked on Jan 30, 2023

Google ScholarTM

Check


Items in GCRIS Repository are protected by copyright, with all rights reserved, unless otherwise indicated.