The Community for Technology Leaders
RSS Icon
Subscribe
Issue No.07 - July (2011 vol.22)
pp: 1135-1141
Jukka Antikainen , University of Eastern Finland, Joensuu
Jiří Havel , Brno University of Technology, Brno
Radovan Jošth , Brno University of Technology, Brno
Adam Herout , Brno University of Technology, Brno
Pavel Zemčík , Brno University of Technology, Brno
Markku Hauta-Kasari , University of Eastern Finland, Joensuu
ABSTRACT
This article presents an optimized algorithm for Nonnegative Tensor Factorization (NTF), implemented in the CUDA (Compute Uniform Device Architecture) framework, that runs on contemporary graphics processors and exploits their massive parallelism. The NTF implementation is primarily targeted for analysis of high-dimensional spectral images, including dimensionality reduction, feature extraction, and other tasks related to spectral imaging; however, the algorithm and its implementation are not limited to spectral imaging. The speedups measured on real spectral images are around 60-100{\times} compared to a traditional C implementation compiled with an optimizing compiler. Since common problems in the field of spectral imaging may take hours on a state-of-the-art CPU, the speedup achieved using a graphics card is attractive. The implementation is publicly available in the form of a dynamically linked library, including an interface to MATLAB, and thus may be of help to researchers and engineers using NTF on large problems.
INDEX TERMS
Nonnegative tensor factorization, spectral analysis, GPU.
CITATION
Jukka Antikainen, Jiří Havel, Radovan Jošth, Adam Herout, Pavel Zemčík, Markku Hauta-Kasari, "Nonnegative Tensor Factorization Accelerated Using GPGPU", IEEE Transactions on Parallel & Distributed Systems, vol.22, no. 7, pp. 1135-1141, July 2011, doi:10.1109/TPDS.2010.194
REFERENCES
[1] A. Andriyashin, J. Parkkinen, and T. Jaaskelainen, "Illuminant Dependence of PCA, NMF and NTF in Spectral Color Imaging," Proc. 19th Int'l Conf. Pattern Recognition (ICPR '08), pp. 1-4, Dec. 2008.
[2] M.W. Berry, M. Browne, A.N. Langville, P.V. Pauca, and R.J. Plemmons, "Algorithms and Applications for Approximate Nonnegative Matrix Factorization," Computational Statistics & Data Analysis, vol. 52, no. 1, pp. 155-173, Sept. 2007.
[3] A. Cichocki and S.-I. Amari, Adaptive Blind Signal and Image Processing, first ed. Wiley, June 2002.
[4] A. Cichocki, R. Zdunek, S. Choi, R. Plemmons, and S.I. Amari, Novel Multi-Layer Non-Negative Tensor Factorization with Sparsity Constraints, pp. 271-280. Springer, 2007.
[5] L. Grippo and M. Sciandrone, "On the Convergence of the Block Nonlinear Gauss-Seidel Method under Convex Constraints," Operations Research Letters, vol. 26, no. 3, pp. 127-136, 2000.
[6] Khronos OpenCL Working Group. "The OpenCL Specification," http://www.khronos.org/registry/cl/specs opencl-1.0.48.pdf, Oct. 2009.
[7] O. Haavisto, Reflectance Spectrum Analysis of Mineral Flotation Froths and Slurries. PhD thesis, Helsinki Univ. of Tech nology, 2009.
[8] M. Harris, "Optimizing Parallel Reduction in CUDA," nVidia, http://developer.download.nvidia.com/compute/ cuda/1_1/Website/projects/reduction/ docreduction.pdf, Jan. 2010.
[9] T. Hazan, S. Polak, and A. Shashua, "Sparse Image Coding Using a 3D Non-Negative Tensor Factorization," Proc. 10th IEEE Int'l Conf. Computer Vision (ICCV '05), vol. 1, pp. 50-57, Oct. 2005.
[10] A. Kaarna, A. Andriyashin, S. Nakauchi, and J. Parkkinen, Multiresolution Approach in Computing NTF, pp. 334-343. Springer, 2007.
[11] M. Kim and S. Choi, "Monaural Music Source Separation: Nonnegativity, Sparseness, and Shift-Invariance," Independent Component Analysis and Blind Signal Separation, J.P. Rosca, D. Erdogmus, J.C. Príncipe, and S. Haykin, eds., pp. 617-624, Springer, 2006.
[12] M. Mørup, L.K. Hansen, C.S. Herrmann, J. Parnas, and S.M. Arnfred, "Parallel Factor Analysis as an Exploratory Tool for Wavelet Transformed Event-Related EEG," NeuroImage, vol. 29, no. 3, pp. 938-947, Feb. 2006.
[13] K. Nishino, M. Nakamura, M. Matsumoto, O. Tanno, and S. Nakauchi, "Imaging of Cosmetics Foundation Distribution by a Spectra Difference Enhancement Filter," Proc. Fifth European Conf. Colour in Graphics, Imaging, and Vision (CGIV), pp. 275-281, June 2010.
[14] NVIDIA. NVIDIA CUDA Programming Guide 2.0. 2008.
[15] J.B. Reeves, G.W. McCarty, D.W. Rutherford, and R.L. Wershaw, "Mid-Infrared Diffuse Reflectance Spectroscopic Examination of Charred Pine Wood, Bark, Cellulose, and Lignin: Implications for the Quantitative Determination of Charcoal in Soils," Applied Spectroscopy, vol. 62, no. 2, pp. 182-189, 2008.
[16] A. Smilde, R. Bro, and P. Geladi, Multi-Way Analysis: Applications in the Chemical Sciences. John Wiley and Sons, 2004.
[17] T. Yotsuda, T. Yamamoto, H. Ishizawa, T. Nishimatsu, and E. Toba, "Near Infrared Spectral Imaging for Water Absorbency of Woven Fabrics," Proc. 21st IEEE Instrumentation and Measurement Technology Conf. (IMTC '04), vol. 3, pp. 2258-2261, May 2004.
[18] Q. Zhang, M.W. Berry, B.T. Lamb, and T. Samuel, A Parallel Nonnegative Tensor Factorization Algorithm for Mining Global Climate Data, pp. 405-415. Springer, 2009.
15 ms
(Ver 2.0)

Marketing Automation Platform Marketing Automation Tool