Hyper-parameter learning for graph based semi-supervised learning algorithms

Abstract

Semi-supervised learning algorithms have been successfully applied in many applications with scarce labeled data, by utilizing the unlabeled data. One important category is graph-based semi-supervised learning algorithms, for which the performance depends considerably on the quality of the graph, or its hyperparameters. In this thesis, we deal with the less explored problem of learning the graphs. We propose a graph learning method for the harmonic energy minimization method; this is done by minimizing the leave-one-out prediction error on labeled data points. We use a gradient-based method and designed an efficient algorithm which significantly accelerates the calculation of the gradient by applying the matrix inversion lemma and using careful pre-computation. We also propose several novel approaches for graph learning regularization, which is so far a less explored field as well. Experimental results show that the learning method is effective in improving the performance of the method.