Chih-Chung Chang, Chih-Wei Hsu, and Chih-Jen Lin (1999)
The analysis of decomposition methods for support vector machines
In: Proceeding of IJCAI99, SVM workshop.
The support vector machine generally requires the solution of a large dense quadratic programming problem. Up to now, very few methods can handle the memory problem and an important one is the ``decomposition method.'' However, there is no convergence proof so far. In this paper, we connect this method to projected gradient methods and provide theoretical proofs for a version of decomposition methods.