NeuroCOLT

Neural Networks and Computational Learning Theory

 

About NeuroCOLT

Papers Archive

1994 1995
1996 1997
1998 1999
2000 2001
2002

Books

info@neurocolt.org

NeuroCOLT Technical Report NC-TR-02-129


2002-129
Provably Fast Training Algorithms for Support Vector Machines
Jose Balcazar, Yang Dai, Junichi Tanaka and Osamu Watanabe


ABSTRACT

Support Vector Machines are a family of algorithms for the analysis of data based on convex Quadratic Programming. We focus on their use for classification, where the SVM algorithms work by maximizing the margin of a classifying hyperplane in a feature space. In this paper, based on a variation of Random Sampling Techniques, techniques successfully used for similar problems, we derive a randomized algorithm for training SVMs and formally prove an upper bound on the expected running time
which is quasilinear with respect to the number of data points. To our knowledge, this is the first algorithm with a quasilinear bound that can handle SVMs with kernels.

 


Download Postscript