Improving the efficiency of IRWLS SVMs using parallel Cholesky factorization Articles
Overview
published in
- PATTERN RECOGNITION LETTERS Journal
publication date
- December 2016
start page
- 91
end page
- 98
volume
- 84
Digital Object Identifier (DOI)
International Standard Serial Number (ISSN)
- 0167-8655
Electronic International Standard Serial Number (EISSN)
- 1872-7344
abstract
- This paper proposes a new and efficient parallel schema of the Iterative Re-Weighted Least Squares (IRWLS) procedure to solve Support Vector Machines (SVMs). This procedure makes use of a parallel Cholesky decomposition to solve in every iteration the linear systems. In particular, we provide two different solutions, a parallel implementation of the IRWLS procedure (PIRWLS) to solve a full SVM and a new parallel implementation of a semi-parametric model of SVM (PSIRWLS). Both solutions have been implemented for multicore and multiprocessor environments with shared memory. We have benchmarked these algorithms against LibSVM, SVMLight and PS-SVM. Experimental results show that using large datasets, our systems offer better parallelization capabilities and higher speed. (C) 2016 Elsevier B.V. All rights reserved.
Classification
keywords
- support vector machines; multicore; cholesky factorization; support vector machine; classifiers; compact