A recurrent neural network implementation using the graphics processing unit

Graduate Thesis Or Dissertation

Public Deposited

Citeable URL: https://ir.library.oregonstate.edu/concern/graduate_thesis_or_dissertations/9g54xn93x

Descriptions

Attribute Name	Values
Creator	Moore, Christopher
Abstract	We took the back-propagation algorithms of Werbos for recurrent and feed-forward neural networks and implemented them on machines with graphics processing units (GPU). The parallelism of these units gave our implementations a 10 to 100 fold increase in speed. For nets with less than 20 neurons the machine performed faster without using the GPU, but for larger nets use of the GPU always gave better times.
License	All rights reserved
Resource Type	Masters Thesis
Date Available	2009-12-17T16:37:08+00:00
Date Issued	2009-10-23
Degree Level	Master's
Degree Name	Master of Science (M.S.)
Degree Field	Computer Science
Degree Grantor	Oregon State University
Commencement Year	2010
Advisor	Bailey, Mike
Committee Member	Cull, Paul Wright, Dawn Metoyer, Ron
Academic Affiliation	Electrical Engineering and Computer Science
Non-Academic Affiliation	Oregon State University. Graduate School
Subject	Neural networks (Computer science) Rendering (Computer graphics)
Rights Statement	In Copyright
Publisher	Oregon State University
Peer Reviewed	No
Language	English [eng]
Replaces	http://hdl.handle.net/1957/13658

Parents:

This work has no parents.

In Collection:

Thumbnail	Title	Date Uploaded	Visibility	Actions
	MooreChristopherE2009.pdf	2017-10-20	Public	Download