This paper concerns the design of a Support Vector Machine (SVM) appropriate for the learning of Boolean functions. This is motivated by the need of a more sophisticated algorithm for classification in discrete attribute spaces. Classification in discrete attribute spaces is reduced to the problem of learning Boolean functions from examples of its input/output behavior. Since any Boolean functi...

In this paper, we investigate how to scale up kernel methods to take on large-scale problems, on which deep neural networks have been prevailing. To this end, we leverage existing techniques and develop new ones. These techniques include approximating kernel functions with features derived from random projections, parallel training of kernel models with 100 million parameters or more, and new s...

A new solution for the problem of selecting poles of the two-parameter Kautz functions in Volterra models of any order is proposed. The usual large number of parameters required to represent the Volterra kernels can be reduced by describing each kernel using a basis of orthonormal functions, such as the Kautz basis. The resulting model can be truncated into fewer terms if the Kautz functions ar...

Choosing an appropriate kernel is very important and critical when classifying a new problem with Support Vector Machine. So far, more attention has been paid on constructing new kernels and choosing suitable parameter values for a specific kernel function, but less on kernel selection. Furthermore, most of current kernel selection methods focus on seeking a best kernel with the highest classif...

While kernel canonical correlation analysis (kernel CCA) has been applied in many problems, the asymptotic convergence of the functions estimated from a finite sample to the true functions has not yet been established. This paper gives a rigorous proof of the statistical convergence of kernel CCA and a related method (NOCCO), which provides a theoretical justification for these methods. The res...

This work proposes a family of language-independent semantic kernel functions defined for individuals in an ontology. This allows exploiting wellfounded kernel methods for several mining applications related to OWL knowledge bases. Namely, our method integrates the novel kernel functions with a support vector machine that can be set up to work with these representations. In particular, we prese...

We present two classes of random walks restricted to the quarter plane whose generating function is not holonomic. The non-holonomy is established using the iterated kernel method, a recent variant of the kernel method. This adds evidence to a recent conjecture on combinatorial properties of walks with holonomic generating functions. The method also yields an asymptotic expression for the numbe...

The kernel embedding of distributions is a popular machine learning technique to manipulate probability distributions and is an integral part of numerous applications. Its empirical counterpart is an estimate from a finite set of samples from the distribution under consideration. However, for large-scale learning problems the empirical kernel embedding becomes infeasible to compute and approxim...

this paper concentrates on a new procedure which experimentally recognises gears and bearings faults of a typical gearbox system using a least square support vector machine (lssvm). two wavelet selection criteria maximum energy to shannon entropy ratio and maximum relative wavelet energy are used and compared to select an appropriate wavelet for feature extraction. the fault diagnosis method co...