Parametric approach to optimal control

We consider the optimal control problem from view point of parametric aspects. We have examined two cases of the parameterized problems. First case describes the situation when the objective functional contains time t as a parameter. We also show how to apply the parametric optimization techniques, such as pathfollowing methods, for finding a nominal optimal control path.     

A Numerical Approach for Solving Some Convex Maximization Problems

We are concerned with concave programming or the convex maximization problem. In this paper, we propose a method and algorithm for solving the problem which are based on the global optimality conditions first obtained by Strekalovsky (Soviet Mathematical Doklady, 8(1987)). The method continues approaches given in (Journal of global optimization, 8(1996); Journal of Nolinear and convex Analyses 4(1)(2003)). Under certain assumptions a convergence property of the proposed method has been established. Some computational results are reported. Also, it has been shown that the problem of finding the largest eigenvalue can be found by the proposed method.     

Conditional Gradient Method for Double-Convex Fractional Programming Matrix Problems

We consider the problem of optimizing the ratio of two convex functions over a closed and convex set in the space of matrices. This problem appears in several applications and can be classified as a double-convex fractional programming problem. In general, the objective function is nonconvex but, nevertheless, the problem has some special features. Taking advantage of these features, a conditional gradient method is proposed and analyzed, which is suitable for matrix problems. The proposed scheme is applied to two different specific problems, including the well-known trace ratio optimization problem which arises in many engineering and data processing applications. Preliminary numerical experiments are presented to illustrate the properties of the proposed scheme.     

Parametric network utility maximization problem

We show how to solve the parametric utility maximization problem with a continuous parameter in a finite number of steps in order to obtain a solution with given accuracy. Also, we propose a new approach for the discretization of time for the parametric utility maximization problem with Lipschitz utility function. Some numerical results are provided.     

Large margin shortest path routing

A new discriminative approach to routing inspired by the large margin criterion serving as a basis for support vector machines is presented. The proposed formulation uses the benefit of the dualization convex program, and it is possible for standard solvers to learn the weighting metrics of the shortest path routing. In order to demonstrate this and due to its simplicity, the single path flow allocation problem is considered in this article. It was found that the weight settings performed within a few percent of that of the optimal general routing where the flow for each demand was optimally distributed over all paths between the source and destination.