Multiple Kernel Learning Using Regularized Ho-Kashyap Classifier in Empirical Kernel Mapping Space

In this paper, we have done research on Multiple Kernel Learning in Empirical Kernel Mapping Space. We find the combination of kernels in empirical kernel mapping space means weighted fusion of empirical kernel mapping samples. And then, we develop a kind of multiple kernel regularized Ho-Kashyap classifier to realize multiple kernel classification in empirical kernel mapping space. The experimental results on benchmark datasets demonstrate the feasibility and effectiveness of the proposed method in empirical kernel mapping space.