Distributed deep neural networks

dc.contributor.advisorCaramanis, Constantine
dc.contributor.committeeMemberKhurshid, Sarfraz
dc.creatorMullapudi, Subhash Venkat
dc.date.accessioned2018-02-26T17:34:09Z
dc.date.available2018-02-26T17:34:09Z
dc.date.created2017-12
dc.date.issued2017-11-09
dc.date.submittedDecember 2017
dc.date.updated2018-02-26T17:34:09Z
dc.description.abstractDeep neural networks have become popular for solving machine learning problems in the field of computer vision. Although computers have reached parity in the task of image classification in machine learning competitions, the task of mining massive training data often takes expensive hardware a long time to process. Distributed protocol for model training can be attractive because less powerful distributed nodes are cheaper to operate than specialized high-performance cluster. Stochastic gradient descent (SGD) is a popular optimizer at the heart of many deep learning systems. To investigate the performance of distributed asynchronous SGD, Tensorflow deep learning framework was tested with Downpour SGD and Delay Compensated SGD to see effect of model training in typical commercial environments. Experimental results show that both Downpour and Delay Compensated SGD are viable protocols for distributed deep learning.
dc.description.departmentElectrical and Computer Engineering
dc.format.mimetypeapplication/pdf
dc.identifierdoi:10.15781/T23B5WR2X
dc.identifier.urihttp://hdl.handle.net/2152/63752
dc.language.isoen
dc.subjectDistributed
dc.subjectDeep neural networks
dc.subjectStochastic gradient descent
dc.subjectSGD
dc.subjectArtificial neural networks
dc.subjectANN
dc.subjectDNN
dc.titleDistributed deep neural networks
dc.typeThesis
dc.type.materialtext
thesis.degree.departmentElectrical and Computer Engineering
thesis.degree.disciplineElectrical and Computer Engineering
thesis.degree.grantorThe University of Texas at Austin
thesis.degree.levelMasters
thesis.degree.nameMaster of Science in Engineering

Access full-text files

Original bundle

Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
MULLAPUDI-MASTERSREPORT-2017.pdf
Size:
939.17 KB
Format:
Adobe Portable Document Format

License bundle

Now showing 1 - 1 of 1
No Thumbnail Available
Name:
LICENSE.txt
Size:
1.85 KB
Format:
Plain Text
Description: