Multiple Parameter Selection for LS-SVM Using Smooth Leave-One-Out Error

In least squares support vector (LS-SVM), the key challenge lies in the selection of free parameters such as kernel parameters and tradeoff parameter. However, when a large number of free parameters are involved in LS-SVM, the commonly used grid search method for model selection is intractable. In this paper, SLOO-MPS is proposed for tuning multiple parameters for LS-SVM to overcome this problem. This method is based on optimizing the smooth leaveone-out error via a gradient descent algorithm and feasible to compute. Extensive empirical comparisons confirm the feasibility and validation of the SLOO-MPS.

• Type: Conference
• Year: 2005
• Where: ISNN
• Authors: Liefeng Bo, Ling Wang, Licheng Jiao
• Keywords: Free Parameters, ISNN 2005, Kernel Parameters, Parameters