The complexity of a special convex programming problem connected with nonlinear optimization

In this paper, an algorithm for solving a special convex programming problem (CPP) is proposed. We consider a CPP with an objective function whose values and gradients are not easy to calculate. Algorithm for infinite feasible sets is developed.     

Direct theorems in semi-infinite convex programming

We show that a semi-infinite quasi-convex program with certain regularity conditions possesses finitely constrained subprograms with the same optimal value. This result is applied to various problems.Research partially funded by NSERC A4493.     

On duality theory of convex semi-infinite programming

In this article we discuss weak and strong duality properties of convex semi-infinite programming problems. We use a unified framework by writing the corresponding constraints in a form of cone inclusions. The consequent analysis is based on the conjugate duality approach of embedding the problem into a parametric family of problems parameterized by a finite-dimensional vector.     

Optimality conditions for nondifferentiable convex semi-infinite programming

This paper gives characterizations of optimal solutions to the nondifferentiable convex semi-infinite programming problem, which involve the notion of Lagrangian saddlepoint. With the aim of giving the necessary conditions for optimality, local and global constraint qualifications are established. These constraint qualifications are based on the property of Farkas-Minkowski, which plays an important role in relation to certain systems obtained by linearizing the feasible set. It is proved that Slater's qualification implies those qualifications.     

On solving convex quadratic semi-infinite programming probelms

In this paper we study a from of convex quadratic semi-infinite programming problems with finitely many variables and infinitely many constraints over a compact metric space. An entropic path-following algorithum is introduced with a convergence proof. Some practical implementations and numerical experiments are also included     

Posynomial geometric programming as a special case of semi-infinite linear programming

This paper develops a wholly linear formulation of the posynomial geometric programming problem. It is shown that the primal geometric programming problem is equivalent to a semi-infinite linear program, and the dual problem is equivalent to a generalized linear program. Furthermore, the duality results that are available for the traditionally defined primal-dual pair are readily obtained from the duality theory for semi-infinite linear programs. It is also shown that two efficient algorithms (one primal based and the other dual based) for geometric programming actually operate on the semi-infinite linear program and its dual.     

Best proximity point theorems: unriddling a special nonlinear programming problem

This paper addresses the non-linear programming problem of globally minimizing the real valued function x?d(x,Sx) where S is a generalized proximal contraction in the setting of a metric space. Eventually, one can obtain optimal approximate solutions to some fixed-point equations in the event that they have no solution.     

Directional differentiability of the optimal value function in convex semi-infinite programming

In this paper, directional differentiability properties of the optimal value function of a parameterized semi-infinite programming problem are studied. It is shown that if the unperturbed semi-infinite programming problem is convex, then the corresponding optimal value function is directionally differentiable under mild regularity assumptions. A max-min formula for the directional derivatives, well-known in the finite convex case, is given.     

Nonsmooth semi-infinite programming problem using Limiting subdifferentials

In this paper, we establish necessary and sufficient optimality conditions for nonsmooth semi-infinite programming problem using the powerful tool of limiting subdifferentials. We also formulate Wolfe and Mond-Weir type duals for nonsmooth semi-infinite programming problem and establish weak, strong and strict converse duality theorems for semi-infinite programming problem and the corresponding dual problems.     

A sequential quadratic programming with a dual parametrization approach to nonlinear semi-infinite programming

In the sequel of the work reported in Liu et al. (1999), in which a method based on a dual parametrization is used to solve linear-quadratic semi-infinite programming (SIP) problems, a sequential quadratic programming technique is proposed to solve nonlinear SIP problems. A merit function to measure progress toward the solution and a procedure to compute the penalty parameter are also proposed.     

Duality-based regularization in a linear convex mathematical programming problem

For a linear convex mathematical programming (MP) problem with equality and inequality constraints in a Hilbert space, a dual-type algorithm is constructed that is stable with respect to input data errors. In the algorithm, the dual of the original optimization problem is solved directly on the basis of Tikhonov regularization. It is shown that the necessary optimality conditions in the original MP problem are derived in a natural manner by using dual regularization in conjunction with the constructive generation of a minimizing sequence. An iterative regularization of the dual algorithm is considered. A stopping rule for the iteration process is presented in the case of a finite fixed error in the input data.     

Path-following proximal approach for solving Ill-posed convex semi-infinite programming problems

For a class of ill-posed, convex semi-infinite programming problems, a regularized path-following strategy is developed. This approach consists in a coordinated application of adaptive discretization and prox-regularization procedures combined with a penalty method. At each iteration, only an approximate minimum of a strongly convex differentiable function has to be calculated, and this can be done by any fast-convergent algorithm. The use of prox-regularization ensures the convergence of the iterates to some solution of the original problem. Due to regularization, an efficient deleting rule is applicable, which excludes an essential part of the constraints in the discretized problems.This research was supported by the German Research Society (DFG).The authors are grateful to the anonymous referees for their valuable comments.     

Feasible partition problem in reverse convex and convex mixed-integer programming

In this paper we consider the consistent partition problem in reverse convex and convex mixed-integer programming. In particular we will show that for the considered classes of convex functions, both integer and relaxed systems can be partitioned into two disjoint subsystems, each of which is consistent and defines an unbounded region. The polynomial time algorithm to generate the partition will be proposed and the algorithm for a maximal partition will also be provided.     

A comparative study of several semi-infinite nonlinear programming algorithms

A computational comparison of several recent semi-infinite nonlinear programming algorithms is presented. Because of the importance of this area, there have recently been proposed a number of algorithms having a global convergence property. The algorithms used in this study include a feasible direction method and some variants of successive quadratic programming methods. Robustness and relative efficiency of the algorithms are examined on the basis of the results for some standard test examples chosen from earlier literature.     

Duality gaps in semi-infinite linear programming—an approximation problem

This paper builds upon the relationship between the objective function of a semi-infinite linear program and its constraints to identify a class of semi-infinite linear programs which do not have a duality gap. The key idea is to guarantee the approximation of the primal program by a sequence of linear programs where thenth approximating program is to minimize the objective function subject to the firstn constraints. The paper goes on to show that any program not in the identified class can be linearly perturbed into it with the optimal value of the perturbed program converging to the optimal value of the original program. The results are then extended to the case when an uncountable number of constraints are present by reducing this case to the countable case.