This paper presents an alternating minimization algorithm used to train radial basis function networks. The algorithm is a modiication of an interior point method used in solving primal linear programs. The resulting algorithm is shown to have a convergence rate on the order of p nL iterations where n is a measure of the network size and L is a measure of the resulting solution's accuracy.

Large-scale optimization problems that seek sparse solutions have become ubiquitous. They are routinely solved with various specialized first-order methods. Although such methods often fast, they usually struggle not-so-well-conditioned problems. In this paper, variants of an interior point-proximal method multipliers proposed and analyzed for class. Computational experience on a variety proble...

We propose a new interior point based method to minimize a linear function of a matrix variable subject to linear equality and inequality constraints over the set of positive semideenite matrices. We present a theoretical convergence analysis, and show that the approach is very eecient for graph bisection problems, such as max-cut. The approach can also be applied to max-min eigenvalue problems.

‎A full Nesterov-Todd (NT) step infeasible interior-point algorithm‎ ‎is proposed for solving monotone linear complementarity problems‎ ‎over symmetric cones by using Euclidean Jordan algebra‎. ‎Two types of‎ ‎full NT-steps are used‎, ‎feasibility steps and centering steps‎. ‎The‎ ‎algorithm starts from strictly feasible iterates of a perturbed‎ ‎problem‎, ‎and, using the central path and feasi...

In this paper the multi-valued versions of some well-known hybrid fixed point theorems of Dhage [6, 7] in Banach algebras are proved. As an application, an existence theorem for a certain integral inclusion in Banach algebras is proved.