On the duality of a non-convex optimization problem and the strong unicity constant in linear Chebyshev approximation (Q1192138)

A semi-infinite maximization problem with a quasi-convex upper semicontinuous objective function (point-valued) and linear constraints is considered. To this primal problem a dual problem is introduced and weak and strong duality theorems are provided. The results can be applied to the computation of the strong unicity constant in linear Chebyshev approximation. An algorithm to compute this constant is shown to converge and the use of it is illustrated with an example.