Semi-supervised classification based on p-norm multiple kernel learning with manifold regularization
School of Automation and Electrical Engineering, University of Science & Technology Beijing, Tao Yang, Dongmei Fu, School of Automation and Electrical Engineering, University of Science & Technology Beijing · Journal of Systems Engineering and Electronics · 2016
Consider the efficiency of p-norm multiple kernel learning (MKL), which is extended to a semi-supervised learning (SSL) scenario by applying the manifold regularization technique. A manifold regularized p-norm multiple kernels model is constructed and applied to a semi-supervised classification task. Solutions are proposed for the case of p = 1, p > 1 and p = ∞, with an analysis of theorems and their proofs. In addition, experiments are conducted on several datasets using state-of-the-art methods to verify the efficiency of the proposed manifold regularized p-norm multiple kernels model in semi-supervised classification.