Nonlinear discrete optimization. An algorithmic theory (Q993280): Difference between revisions

The main purpose of this monograph is to describe in a unified manner several algorithmic aspects concerning nonlinear discrete optimization, in general by extending the well-established theory of linear discrete optimization. The book consists of 6 chapters containing both theoretical aspects and applications. The main part of the monograph can be divided into three parts: {\parindent=5mm \begin{itemize}\item[-] Chapter 2 on convex discrete maximization. The methods used in this chapter are mostly geometric, several polynomial time algorithms being provided for convex maximization in various situations. There are presented as well some direct applications including: quadratic binary programming, matroids and vector partitioning problems. \item[-] Chapters 3--5 on nonlinear integer programming. The main subjects of these chapters are the introduction of a Graver basis of an integer matrix and the theory of \(n\)-fold integer programming. Several applications of this powerful theory are presented in Chapter 5 including multi-index transportation problems, privacy in statistical databases, and extensions to hierarchical margins. \item[-] Chapter 6 on nonlinear combinatorial optimization. Various polynomial time methods are provided for several combinatorial structures. The concrete applications include nonlinear bipartite matching and experimental design. \end{itemize}} The book is well written and addressed to applied mathematicians, engineers and scientists.