Support Vector Machines (SVMs) perform pattern recognition between two point classes by nding a decision surface determined by certain points of the training set, termed Support Vectors (SV). This surface, which in some feature space of possibly innnite dimension can be regarded as a hyperplane, is obtained from the solution of a problem of quadratic programming that depends on a regularization...

Refined concepts, such as Rademacher estimates of model complexity and nonlinear criteria for weighting empirical classification errors, represent recent and promising approaches to characterize the generalization ability of Support Vector Machines (SVMs). The advantages of those techniques lie in both improving the SVM representation ability and yielding tighter generalization bounds. On the o...

In this paper, we propose some improvements for the implementations of least squares support vector machine classifiers (LS-SVM). An improved conjugate gradient scheme is proposed for solving the optimization problems in LS-SVM, and an improved SMO algorithm is put forward for the general unconstrained quadratic programming problems which is the case of LS-SVM without the bias term. Numerical e...

Introduction: The ability of quadratic dose protocol to maintain a good quality image for an overweight and obese patient is well reported. However, a practical approach to the implementation of this protocol in whole-body imaging in Malaysia is currently lacking. Hence, the aim of this study is to derive the quadratic dose formula that suits our PET system. Metho...

In this paper, we study the problem of minimizing the ratio of two quadratic functions subject to a quadratic constraint. First we introduce a parametric equivalent of the problem. Then a bisection and a generalized Newton-based method algorithms are presented to solve it. In order to solve the quadratically constrained quadratic minimization problem within both algorithms, a semidefinite optim...

In this paper, we solve the quadratic $alpha$-functional equations $2f(x) + 2f(y) = f(x + y) + alpha^{-2}f(alpha(x-y)); (0.1)$ where $alpha$ is a fixed non-Archimedean number with $alpha^{-2}neq 3$. Using the fixed point method and the direct method, we prove the Hyers-Ulam stability of the quadratic $alpha$-functional equation (0.1) in non-Archimedean Banach spaces.

We construct a class of codes of length such that the minimum distance outside of a certain subcode is, up to a constant factor, bounded below by the square root of , a well-known property of quadratic residue codes. The construction, using the group algebra of an Abelian group and a special partition or splitting of the group, yields quadratic residue codes, duadic codes, and their generalizat...

1 Quadratic Congruences to Prime Moduli . . . . . . . . . . . . . . . . . . . . . . . . 1 2 Quadratic Congruences to Composite Moduli . . . . . . . . . . . . . . . . . . . . . 5 3 Some Sums of Legendre’s symbols . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 4 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 5 Solutions . . . . . . . . . . . . ...