Deakin home > Deakin University Library > Deakin Research Online > Computing of high breakdown regression estimators without sorting on graphics processing units

Computing of high breakdown regression estimators without sorting on graphics processing units

Beliakov, G., Johnstone, M. and Nahavandi, S. 2012, Computing of high breakdown regression estimators without sorting on graphics processing units, Computing, vol. 94, no. 5, pp. 433-447.

Attached Files (Some files may be inaccessible until you login with your Deakin Research Online credentials)
Name Description MIMEType Size Downloads

Title Computing of high breakdown regression estimators without sorting on graphics processing units
Author(s) Beliakov, G.
Johnstone, M.
Nahavandi, S.
Journal name Computing
Volume number 94
Issue number 5
Start page 433
End page 447
Total pages 15
Publisher Springer-Verlag
Place of publication Wien, Austria
Publication date 2012
ISSN 0010-485X
Keyword(s) cutting plane
GPU
median
order statistic
robust regression
sorting
Summary We present an approach to computing high-breakdown regression estimators in parallel on graphics processing units (GPU).We show that sorting the residuals is not necessary, and it can be substituted by calculating the median. We present and compare various methods to calculate the median and order statistics on GPUs. We introduce an alternative method based on the optimization of a convex function, and showits numerical superiority when calculating the order statistics of very large arrays on GPUs.
Language eng
Field of Research 109999 Technology not elsewhere classified
Socio Economic Objective 970110 Expanding Knowledge in Technology
HERDC Research category C1 Refereed article in a scholarly journal
Copyright notice ©2012, Spinger-Verlag
Persistent URL http://hdl.handle.net/10536/DRO/DU:30046955

Connect to link resolver
 
Unless expressly stated otherwise, the copyright for items in Deakin Research Online is owned by the author, with all rights reserved.

Versions
Version Filter Type
Citation counts: TR Web of Science Citation Count  Cited 2 times in TR Web of Science
Scopus Citation Count Cited 2 times in Scopus
Access Statistics: 46 Abstract Views, 6 File Downloads  -  Detailed Statistics
Created: Mon, 13 Aug 2012, 12:53:56 EST