A novel descriptor for pattern recognition is proposed by using dual-tree complex wavelet features and SVM. The approximate shiftinvariant property of the dual-tree complex wavelet and its good directional selectivity in 2D make it a very appealing choice for pattern recognition. Recently, SVM has been shown to be very successful in pattern recognition. By combining these two tools we find that...

Extreme learning machines are fast models which almost compare to standard SVMs in terms of accuracy, but are much faster. However, they optimise a sum of squared errors whereas SVMs are maximum-margin classifiers. This paper proposes to merge both approaches by defining a new kernel. This kernel is computed by the first layer of an extreme learning machine and used to train a SVM. Experiments ...

We consider a multiscale approximation scheme at scattered sites for functions in Sobolev spaces on the unit sphere Sn. The approximation is constructed using a sequence of scaled, compactly supported radial basis functions restricted to Sn. A convergence theorem for the scheme is proved, and the condition number of the linear system is shown to stay bounded by a constant from level to level, t...

Statistical machine learning plays an important role in modern statistics and computer science. One main goal of statistical machine learning is to provide universally consistent algorithms, i.e., the estimator converges in probability or in some stronger sense to the Bayes risk or to the Bayes decision function. Kernel methods based on minimizing the regularized risk over a reproducing kernel ...

Abstract. Within the conventional framework of a native space structure, a smooth kernel generates a small native space, and “radial basis functions” stemming from the smooth kernel are intended to approximate only functions from this small native space. Therefore their approximation power is quite limited. Recently, Narcowich, Schaback and Ward [NSW], and Narcowich and Ward [NW], respectively,...

This contribution discusses the construction of kernel-based adaptive particle methods for numerical flow simulation, where the finite volume particle method (FVPM) is used as a prototype. In the FVPM, scattered data approximation algorithms are required in the recovery step of the WENO reconstruction. We first show how kernel-based approximation schemes can be used in the recovery step of part...

We propose in this paper a new kernel for time series on structured data in the dynamic time warping family. We demonstrate using the theory of infinitely divisible kernels that this kernel is positive definite, that it is a radial basis kernel and that it reduces to a product kernel when comparing two sequences of the same length. Finally we compare this kernel with the global alignment kernel...

• Many kernel-based learning algorithms have the computational load. • The Nyström low-rank approximation is designed for reducing the computation. • We propose the spectrum decomposition condition with a theoretical justification. • Asymptotic error bounds on eigenvalues and eigenvectors are derived. • Numerical experiments are provided for covariance kernel and Wishart matrix. AMS subject cla...

This paper shows a method to diagnose potential mispronunciations in second language learning by studying the characteristics of the speech produced by a group of native speakers and the speech produced by various non-native groups of speakers from diverse language backgrounds. The method compares the native auditory perception and the non-native spectral representation on the phoneme level usi...

We present probabilistic models which are suitable for class conditional density estimation and can be regarded as shared kernel models where sharing means that each kernel may contribute to the estimation of the conditional densities of an classes. We first propose a model that constitutes an adaptation of the classical radial basis function (RBF) network (with full sharing of kernels among cl...