On GPU–CUDA as preprocessing of fuzzy-rough data reduction by means of singular value decomposition

Data reduction algorithms often produce inaccurate results for loss of relevant information. Recently, the singular value decomposition (SVD) method has been used as preprocessing method in order to deal with high-dimensional data and achieve fuzzy-rough reduct convergence on higher dimensional datasets. Despite the well-known fact that SVD offers attractive properties, its high computational cost remains a critical issue. In this work, we present a parallel implementation of the SVD algorithm on graphics processing units using CUDA programming model. Our approach is based on an iterative parallel version of the QR factorization by means of Givens rotations using the Sameh and Kuck scheme. Our results show significant improvements in terms of performances with respect to the CPU version that encourage its usability for this expensive processing of data.

[1]  Salvatore Cuomo,et al.  A GPU parallel implementation of the Local Principal Component Analysis overcomplete method for DW image denoising , 2016, 2016 IEEE Symposium on Computers and Communication (ISCC).

[2]  H. Andrews,et al.  Singular value decompositions and digital image processing , 1976 .

[3]  Salvatore Cuomo,et al.  Local principal component analysis overcomplete method: A GPU parallel implementation combining shared and global memories , 2016, 2016 International Conference on High Performance Computing & Simulation (HPCS).

[4]  Qiang Shen,et al.  Fuzzy-rough data reduction with ant colony optimization , 2005, Fuzzy Sets Syst..

[5]  Rama Devi,et al.  Fuzzy Rough Data Reduction Using SVD , 2011 .

[6]  Salvatore Cuomo,et al.  Toward a Multi-level Parallel Framework on GPU Cluster with PetSC-CUDA for PDE-based Optical Flow Computation , 2015, ICCS.

[7]  Salvatore Cuomo,et al.  GPU Profiling of Singular Value Decomposition in OLPCA Method for Image Denoising , 2016, 3PGCIC.

[8]  S T Roweis,et al.  Nonlinear dimensionality reduction by locally linear embedding. , 2000, Science.

[9]  Qiang Shen,et al.  A modular approach to generating fuzzy rules with reduced attributes for the monitoring of complex systems , 2000 .

[10]  Mohamed Quafafou,et al.  Generalized rough sets based feature selection , 2000, Intell. Data Anal..

[11]  Salvatore Cuomo,et al.  A GPU Algorithm in a Distributed Computing System for 3D MRI Denoising , 2015, 2015 10th International Conference on P2P, Parallel, Grid, Cloud and Internet Computing (3PGCIC).

[12]  Qiang Shen,et al.  Fuzzy-rough sets for descriptive dimensionality reduction , 2002, 2002 IEEE World Congress on Computational Intelligence. 2002 IEEE International Conference on Fuzzy Systems. FUZZ-IEEE'02. Proceedings (Cat. No.02CH37291).

[13]  Pavel Pudil,et al.  Novel Methods for Feature Subset Selection with Respect to Problem Knowledge , 1998 .

[14]  David J. Kuck,et al.  On Stable Parallel Linear System Solvers , 1978, JACM.

[15]  Salvatore Cuomo,et al.  A parallel PDE-based numerical algorithm for computing the Optical Flow in hybrid systems , 2017, J. Comput. Sci..

[16]  Livia Marcellino,et al.  Deconvolution of 3D Fluorescence Microscopy Images Using Graphics Processing Units , 2011, PPAM.

[17]  Gene H. Golub,et al.  Singular value decomposition and least squares solutions , 1970, Milestones in Matrix Computation.