A recurrent neural network-based method for training probabilistic Support Vector Machine
Hadi Sadoghi Yazdi, Sohrab Effati, Zahra Saberi · International Journal of Signal and Imaging Systems Engineering · 2009
In this paper, Support Vector Machine (SVM) is reformulated to a recurrent neural network model which can be described by the nonlinear dynamic system. In the proposed algorithm, an iterative training procedure is proposed independent of initial point. Also probabilistic constraints are recommended for reducing effect of noisy samples in training procedure and appearance of incorrect Support Vectors (SV). Probabilistic constraints admit using knowledge about distribution function of samples. A set of differential equations is used to modelling of the proposed probabilistic SVM. These equations are converged to optimal solution for SVM. The Euler method is used to solve differential equation. The primal and dual problem of SVM is solved by this model. Enough information is given for finding optimal hyper plane. Capability of the proposed method is shown by experimental results in the Optical Character Recognition (OCR) and synthetic data.