In Chapter , the formulation of SVM concluded that SVM can be trained by solving Eqn. , subject to the constraints in Eqn. . This is standard constrained Quadratic Programming. In this chapter, the practical problems in solving this QP problem will be discussed. The effects of using different solvers are shown in Section . Section and show two major approaches that make the training of large classification problems tractable. In Section , the complexity and scalability of SVM training are discussed.

- Solving the Quadratic Programming Problem
- Decomposition Algorithm for the SVM training
- Incremental Training of the Support Vector Machine
- Complexity and Scalability of the Training Algorithm

Thu Sep 10 11:05:30 BST 1998