A new regression technique based on Vapnikâ€™s concept of support vectors is introduced. We compare support vector regression (SVR) with a committee regression technique (bagging) based on regressionâ€¦ (More)

Several decompositions of, symmetric matrices for calculating inertia and solving systems of linear equations are discussed. New partial pivoting strategies for decomposing symmetric matrices areâ€¦ (More)

Since the publication of Shepp and Vadi's [ 14] maximum likelihood reconstruction algorithm for emission tomography (ET), many medical research centers engaged in ET have made an effort to changeâ€¦ (More)

The EM algorithm is the basic approach used to maximize the log likelihood objective function for the reconstruction problem in positron emission tomography (PET). The EM algorithm is a scaledâ€¦ (More)

In this paper we explain some of the changes that have been incorporated in the latest version of the LAPACK subroutine for reducing a symmetric banded matrix to tridiagonal form. These modificationsâ€¦ (More)

This paper persents a program SERRG2 that solves separable elliptic equations on a rectangle. The program uses a matrix decomposition technique to directly solve the linear system arising from aâ€¦ (More)

Most algorithms associated with finding eigenvalues of banded matrices do not have many, if any, obvious vector operations. On vector machines it is not immediately obvious, if sufficient memory isâ€¦ (More)

The QZ algorithm of Moler and Stewart solves the generalized eigenvalue problem of finding and x such that Ax = XBx for real square matrices A and B by first simultaneously reducing A to upperâ€¦ (More)

Maxwellâ€™s equation for modeling the guided waves in a circularly symmetric fiber leads to a family of partial differential equationâ€“eigenvalue systems. Incorporating the boundary condition into aâ€¦ (More)