MINIMIZATION PROBLEM IN MULTIDIMENSIONAL SPACE

摘要

The problem of minimization of improperly stipulated functionals in multidimensional space is very important both theoretically (e.g., it is a relevant sub-procedure in the solution of the so-called inverse problem associated with information, communication and network systems, imaging recognition, etc.) and in an application sense (e.g., in the solution of various physico-chemical and geophysical problems). Leaving aside a detailed review of this issue, we point out only that the family of the minimization techniques may be subdivided conditionally into two groups: the first one includes methods that transform the initial non-convex space into a strict-convex one, while the second group deals with the methods that are adjusted to the specificity of the given non-convex space.