Abstract Details
Activity Number:
|
430
|
Type:
|
Contributed
|
Date/Time:
|
Tuesday, August 6, 2013 : 2:00 PM to 3:50 PM
|
Sponsor:
|
Section on Statistical Computing
|
Abstract - #310261 |
Title:
|
GPUs, Linear Algebra, and Efficient Computing for Gaussian Process Models
|
Author(s):
|
Colin Rundel*+
|
Companies:
|
Duke University
|
Keywords:
|
GPU ;
Gaussian Process ;
Spatial ;
R ;
C++ ;
Computing
|
Abstract:
|
Here we present recent work on implementing R and C++ tools that leverage existing GPU linear algebra libraries to improve the computing performance of common tasks in the implementation of Gaussian process models. We will focus our discussion of areas like covariance calculation (both simple and complex), likelihood evaluation, and prediction - all important computational bottlenecks for even moderately sized data sets. Our implementation builds on existing libraries and allows for the exploitation of commodity GPU hardware without the need for specific expertise in developing for these processors. Finally, we present performance results from a model for large scale wireless sensor network data in order to demonstrate how our approaches are complementary to existing methods for working with large data (e.g. predictive processes).
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2013 program
|
2013 JSM Online Program Home
For information, contact jsm@amstat.org or phone (888) 231-3473.
If you have questions about the Continuing Education program, please contact the Education Department.
The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Copyright © American Statistical Association.