Comparing support vector machines with Gaussian kernels to radial basis function classifiers
Bernhard Schölkopfa b, Kah-Kay Sungc d, Chris J. C. Burgese f, Federico Girosig h, Partha Niyogii j, Tomaso Poggiog h, Vladimir N. Vapnikk l
- aMax-Planck Institut für biologische Kybernetik, Tubingen, Germany
- bMax-Planck-Inst. fur Biol. Kybernetik, Tubingen, Germany
- cDepartment of Information Systems and Computer Science, National University of Singapore, Singapore
- dDept. of Inf. Syst. & Computer Sci., Nat. Univ. of Singapore, Singapore
- eBell Laboratories, Lucent Technologies, Inc., Holmdel, NJ, USA
- f[Bell Laboratories, Lucent Technologies, Inc., Holmdel, NJ, USA]
- gCenter for Biological and Computational Learning, Massachusetts Institute of Technology, Cambridge, MA, USA
- h[Center for Biological and Computational Learning, Massachusetts Institute of Technology, Cambridge, MA, USA]
- iBell Laboratories, Lucent Technologies, Inc., Murray Hill, NJ, USA
- j[Bell Laboratories, Lucent Technologies Inc., Murray Hill, NJ, USA]
- kAT and T Research, Red Bank, NJ, USA
- l
Papers open inside SciRah from the original publisher or repository; SciRah never hosts them.
Abstract
The support vector (SV) machine is a novel type of learning machine, based on statistical learning theory, which contains polynomial classifiers, neural networks, and radial basis function (RBF) networks as special cases. In the RBF case, the SV algorithm automatically determines centers, weights, and threshold that minimize an upper bound on the expected test error. The present study is devoted to an experimental comparison of these machines with a classical approach, where the centers are determined by X-means clustering, and the weights are computed using error backpropagation. We consider three machines, namely, a classical RBF machine, an SV machine with Gaussian kernel, and a hybrid system with the centers determined by the SV method and the weights trained by error backpropagation. Our results show that on the United States postal service database of handwritten digits, the SV machine achieves the highest recognition accuracy, followed by the hybrid system. The SV approach is thus not only theoretically well-founded but also superior in a practical application.
TL;DRThe results show that on the United States postal service database of handwritten digits, the SV machine achieves the highest recognition accuracy, followed by the hybrid system, and the SV approach is thus not only theoretically well-founded but also superior in a practical application. (Semantic Scholar)