Generalizing the bias term of support vector machines
Wenye Li, Kwong-Sak Leung, Kin-Hong Lee · 2007
Based on the study of a generalized form of rep-resenter theorem and a specific trick in construct-ing kernels, a generic learning model is proposed and applied to support vector machines. An algo-rithm is obtained which naturally generalizes the bias term of SVM. Unlike the solution of standard SVM which consists of a linear expansion of ker-nel functions and a bias term, the generalized algo-rithm maps predefined features onto a Hilbert space as well and takes them into special consideration by leaving part of the space unregularized when seek-ing a solution in the space. Empirical evaluations have confirmed the effectiveness from the general-ization in classification tasks. 1