论文信息 - A Common Framework for the Convergence of the GSK, MDM and SMO Algorithms

A Common Framework for the Convergence of the GSK, MDM and SMO Algorithms

Building upon Gilbert's convergence proof of his algorithtm to solve the Minimum Norm Problem, we establish a framework where a much simplified version of his proof allows us to prove the convergence of two algorithms for solving the Nearest Point Problem for disjoint convex hulls, namely the GSK and the MDM algorithms, as well as the convergence of the SMO algorithm for SVMs over linearly separable two-class samples.

José R. Dorronsoro | Jorge López Lázaro

[1] Aapo Hyvärinen,et al. Learning Features by Contrasting Natural Images with Noise , 2009, ICANN.

[2] Václav Hlavác,et al. An iterative algorithm learning the maximal margin classifier , 2003, Pattern Recognit..

[3] Chih-Jen Lin,et al. On the convergence of the decomposition method for support vector machines , 2001, IEEE Trans. Neural Networks.

[4] José R. Dorronsoro,et al. A Simple Proof of the Convergence of the SMO Algorithm for Linearly Separable Problems , 2009, ICANN.

[5] S. Sathiya Keerthi,et al. A fast iterative nearest point algorithm for support vector machine classifier design , 2000, IEEE Trans. Neural Networks Learn. Syst..

[6] E. Gilbert. An Iterative Procedure for Computing the Minimum of a Quadratic Form on a Convex Set , 1966 .

[7] V. N. Malozemov,et al. Finding the Point of a Polyhedron Closest to the Origin , 1974 .