Multi-threaded Support Vector Machines for Pattern Recognition
Support Vector Machines (SVM) have become indispensable tools in the area of pattern recognition. They show powerful classification and regression performance in highly non-linear problems by mapping the input vectors nonlinearly into a high-dimensional feature space through a kernel function. However, the optimization task is numerically expensive since single-threaded implementations are hardly able to cope up with the complex learning task. In this paper, we present a multi-threaded implementation of the Sequential Minimal Optimization (SMO) which reduces the numerical complexity by parallelizing the KKT conditions update, the calculation of the hyperplane offset and the classification task. Our preliminary results both in benchmark datasets and real-world problems show competitive performance to the state-of-the-art tools while the execution running times are considerably faster.
KeywordsSVM OpenMP sequential minimal optimization (SMO)
Unable to display preview. Download preview PDF.
- 2.Asuncion, A., Newman, D.J.: UCI machine learning repository (2010), http://archive.ics.uci.edu/ml/index.html
- 3.Ayat, N., Cheriet, M., Suen, C.: KMOD - a two-parameter SVM kernel for pattern recognition. In: Proc. of ICPR, pp. 30331–30334 (2002)Google Scholar
- 6.Chang, C.C., Lin, C.J.: LIBSVM: A Library for Support Vector Machines. Dep. of Computer Science National Taiwan University, Taipei, Taiwan (2011)Google Scholar
- 8.Cortes, C., Vapnik, V.: Support-vector networks. In: Machine Learning, pp. 273–297 (1995)Google Scholar
- 12.Lopes, N., Correia, D., Pereira, C., Ribeiro, B., Dourado, A.: An Incremental Hypersphere Learning Framework for Protein Membership Prediction. In: Corchado, E., Snášel, V., Abraham, A., Woźniak, M., Graña, M., Cho, S.-B. (eds.) HAIS 2012, Part III. LNCS, vol. 7208, pp. 429–439. Springer, Heidelberg (2012)CrossRefGoogle Scholar
- 13.Platt, J.: Sequential minimal optimization: A fast algorithm for training support vector machines, vol. 208, pp. 1–21. MIT Press (1998)Google Scholar
- 14.Qiao, M., Sung, A.H., Liu, Q.: Feature mining and intelligent computing for MP3 steganalysis. In: Int. Joint Conf. on Bioinformatics, Systems Biology and Intelligent Computing, pp. 627–630. IEEE Computer Society (2009)Google Scholar
- 15.Vapnik, V.: The nature of statistical learning theory. Springer (1995)Google Scholar
- 17.Zhang, R., Wang, W.: Facilitating the applications of support vector machine by using a new kernel. Expert Systems with Applications 38, 14225–14230 (2011)Google Scholar