An interval branch and bound method for global Robust optimization

In this paper, we design a Branch and Bound algorithm based on interval arithmetic to address nonconvex robust optimization problems. This algorithm provides the exact global solution of such difficult problems arising in many real life applications. A code was developed in MatLab and was used to solve some robust nonconvex problems with few variables. This first numerical study shows the interest of this approach providing the global solution of such difficult robust nonconvex optimization problems.

     

Heuristic approaches for support vector machines with the ramp loss

Recently, Support Vector Machines with the ramp loss (RLM) have attracted attention from the computational point of view. In this technical note, we propose two heuristics, the first one based on solving the continuous relaxation of a Mixed Integer Nonlinear formulation of the RLM and the second one based on the training of an SVM classifier on a reduced dataset identified by an integer linear problem. Our computational results illustrate the ability of our heuristics to handle datasets of much larger size than those previously addressed in the literature.     

Solving Nonconvex Planar Location Problems by Finite Dominating Sets

It is well-known that some of the classical location problems with polyhedral gauges can be solved in polynomial time by finding a finite dominating set, i.e. a finite set of candidates guaranteed to contain at least one optimal location.In this paper it is first established that this result holds for a much larger class of problems than currently considered in the literature. The model for which this result can be proven includes, for instance, location problems with attraction and repulsion, and location-allocation problems.Next, it is shown that the approximation of general gauges by polyhedral ones in the objective function of our general model can be analyzed with regard to the subsequent error in the optimal objective value. For the approximation problem two different approaches are described, the sandwich procedure and the greedy algorithm. Both of these approaches lead - for fixed - to polynomial approximation algorithms with accuracy for solving the general model considered in this paper.     

Alternating local search based VNS for linear classification

We consider the linear classification method consisting of separating two sets of points in d-space by a hyperplane. We wish to determine the hyperplane which minimises the sum of distances from all misclassified points to the hyperplane. To this end two local descent methods are developed, one grid-based and one optimisation-theory based, and are embedded into a VNS metaheuristic scheme. Computational results show these approaches to be complementary, leading to a single hybrid VNS strategy which combines both approaches to exploit the strong points of each. Extensive computational tests show that the resulting method can always be expected to approach the global optimum close enough that any deviations from the global optimum are irrelevant with respect to the classification power.     

Minmax-distance approximation and separation problems: geometrical properties

A center hyperplane in the d-dimensional space minimizes the maximum of its distances from a finite set of points A with respect to possibly different gauges. In this note it is shown that a center hyperplane exists which is at (equal) maximum distance from at least d?+?1 points of A. Moreover the projections of the points among these which lie above the center hyperplane cannot be separated by another hyperplane from the projections of those that are below it. When all gauges involved are smooth, all center hyperplanes satisfy these properties. This geometric property allows us to improve and generalize previously existing results, which were only known for the case in which all distances are measured using a common norm. The results also extend to the constrained case where for some points it is prespecified on which side of the hyperplane (above, below or on) they must lie. In this case the number of points lying on the hyperplane plus those at maximum distance is at least d?+?1. It follows that solving such global optimization problems reduces to inspecting a finite set of candidate solutions. Extensions of these results to a separation problem are outlined.     

Improving Interval Analysis Bounds by Translations

We explore how a simple linear change of variable affects the inclusion functions obtained with Interval Analysis methods. Univariate and multivariate polynomial test functions are considered, showing that translation-based methods improve considerably the bounds computed by standard inclusion functions. An Interval Branch-and-Bound method for global optimization is then implemented to compare the different procedures, showing that, although with times higher than those given by Taylor forms, the number of clusters and iterations is strongly reduced.     

Polynomial algorithms for parametric minquantile and maxcovering planar location problems with locational constraints

A location is sought within some convex region of the plane for the central site of some public service to a finite number of demand points. The parametric maxcovering problem consists in finding for eachR>0 the point from which the total weight of the demand points within distanceR is maximal. The parametric minimal quantile problem asks for each percentage α the point minimising the distance necessary for covering demand points of total weight at least α. We investigate the properties of these two closely related problems and derive polynomial algorithms to solve them both in case of either (possibly inflated) Euclidean or polyhedral distances. The research of the first author is partially supported by Grant PB96-1416-C02-02 of Ministerio de Educación y Cultura, Spain.