Additional learning and forgetting by support vector machine and RBF networks

Radial basis function networks (RBFNs) have been widely applied to practical classification problems. In recent years, support vector machines (SVMs) are attracting researchers' interest as promising methods for classification problems. In this paper, we compare those two methods in view of additional learning and forgetting. The authors have reported that the additional learning and active forgetting in RBFNs provide a good performance for classification under the changeable environment. First, a method for additional learning and forgetting in SVMs is proposed. Next, a comparative simulation for a portfolio problems between RBFNs and SVMs is made.