Show simple item record

dc.contributor.advisorBajaj, Chandrajit
dc.creatorHintz, Jeremy James
dc.date.accessioned2017-02-02T15:15:54Z
dc.date.available2017-02-02T15:15:54Z
dc.date.issued2016-12
dc.date.submittedDecember 2016
dc.identifierdoi:10.15781/T24B2X90Q
dc.identifier.urihttp://hdl.handle.net/2152/44614
dc.description.abstractIn this report, we will give a brief overview of selected deep learning technologies in the interest of developing both understanding and motivation for the use of reservoir computing and generative models. Furthermore, we will show that these concepts can be applied to the problem of natural video prediction. Influenced by previous work, we develop a novel architecture called Generative Adversarial Reservoirs (GAR). We use GARs to predict frames of videos from the UCF-101 dataset and show that although some of the quantitative evaluations for our results are below state-of-the-art, utilizing reservoirs allows our model training to converge significantly faster while still achieving qualitatively good results.
dc.format.mimetypeapplication/pdf
dc.language.isoen
dc.subjectDeep learning
dc.subjectReservoir computing
dc.subjectVideo prediction
dc.subjectGenerative Adversarial Networks
dc.titleGenerative Adversarial Reservoirs for natural video prediction
dc.typeThesis
dc.date.updated2017-02-02T15:15:54Z
dc.description.departmentComputational Science, Engineering, and Mathematics
thesis.degree.departmentComputational Science, Engineering, and Mathematics
thesis.degree.disciplineComputational science, engineering, and mathematics
thesis.degree.grantorThe University of Texas at Austin
thesis.degree.levelMasters
thesis.degree.nameMaster of Science in Computational Science, Engineering, and Mathematics
dc.type.materialtext


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record