Interpolation by analytic radial basis functions like the Gaussian and inverse multiquadrics can degenerate in two ways: the radial basis functions can be scaled to become “increasingly flat”, or the data points “coalesce” in the limit while the radial basis functions stays fixed. Both cases call for a careful regularization. If carried out explicitly, this yields a preconditioning technique fo...

It has been shown that Support Vector Machine theory optimizes a smoothness functional hypothesis through kernel application. We present KMOD, a two-parameter SVM kernel with distinctive properties of good discrimination between patterns while preserving the data neighborhood information. In classi£cation problems, the experiments we carried out on the Breast Cancer benchmark produced better pe...

A B-spline kernel combined with RBF is developed, a mixed kernel is obtained. By analyzing the structure of the logging signal characteristics, the method is used to automatically identify the water-flooded status of oilsaturated stratum. The experimental results show that the mixed kernel has high recognition accuracy with the advantages of the short running time.

Our objective is to speed up non-linear SVM prediction while maintaining classification accuracy above an acceptable limit. We generalize Localized Multiple Kernel Learning so as to learn a tree-based primal feature embedding which is high dimensional and sparse. Primal based classification decouples prediction costs from the number of support vectors and our tree-structured features efficientl...

Kernel methods are a powerful set of techniques for learning from data. One of the attractive properties of these techniques is that they rely only on a kernel function which provides the user-defined notion of similarity between two observations, to train the models. This report describes a strategy for evaluating kernel-based predictive models within a cross-validation framework when we also ...

This work proposes a method which enables us to perform kernel Fisher discriminant analysis in the whole eigenspace for face recognition. It employs the ratio of eigenvalues to decompose the entire kernel feature space into two subspaces: a reliable subspace spanned mainly by the facial variation and an unreliable subspace due to finite number of training samples. Eigenvectors are then scaled u...

Kernel methods such as support vector machines (SVMs) have attracted a great deal of popularity in the machine learning and natural language processing (NLP) communities. Polynomial kernel SVMs showed very competitive accuracy in many NLP problems, like part-of-speech tagging and chunking. However, these methods are usually too inefficient to be applied to large dataset and real time purpose. I...

An extension for univariate kernels that deals with missing values is proposed. These extended kernels are shown to be valid Mercer kernels and can adapt to many types of variables, such as categorical or continuous. The proposed kernels are tested against standard RBF kernels in a variety of benchmark problems showing different amounts of missing values and variable types. Our experimental res...