Handbook of semidefinite programming. Theory, algorithms, and applications (Q5915427): Difference between revisions

This handbook is an efficient and well-readable state-of-the-art presentation of one of the most exciting and active research areas in optimization in the last decade. Most of the leading experts in this field contributed to this exhaustive volume and the comprehensive bibliography consists of almost thousand references. The tremendously growing interest in Semidefinite Programming Problems (SDP) in the 1990s was motivated by several applications in different areas and by the development of efficient new algorithms for their solutions. Thus, SDP has attracted researchers from different backgrounds, including mathematicians and engineers from convex programming, linear algebra, numerical optimization, combinatorial optimization, control theory, statistics and others. The book is divided into three parts: Theory, Algorithms as well as Applications and Extensions. Theory: The theory underlying SDP is studied from different points of view. SDP is a straightforward application of general techniques known from convex analysis; it is also an extension of standard linear programming, here, applied over a nonpolyhedral convex cone. Furthermore, chapters of this part deal with duality theory for convex and nonconvex nonlinear cone constrained optimization problems, first and second order optimality conditions, sensitivity analysis, self-dual embeddings as well as robustness in SDP. Algorithms: The development of efficient polynomial-time algorithms at the end of the 1980s initiated an increased research activity in SDP. Now, SDP can be solved in polynomial time by using self-concordant barrier functions. Three chapters deal with interior-point methods: a generalization from methods in linear programming, a class of symmetric primal-dual potential reduction algorithms and primal-dual path-following methods. Eventually, the last chapter in this second part discusses bundle methods, known from nondifferentiable convex programming which are interesting for very large structured SDPs. Applications and Extensions: There exists a broad variety of interesting mathematical and engineering applications of SDP; some of them are considered here: the solution of nonlinear convex relaxation of NP-hard problems in combinatorial optimization, the solution of diverse control problems for which no analytical solutions are known, structural optimization, nonlinear optimization problems with matrix inequality constraints, optimal design of experiments, derivation of bounds in stochastic optimization, matrix completion problems and others. Besides the contributions and bibliography, the editors added an appendix covering most of the main topics of the handbook together with references and links to corresponding chapters. Altogether, this is a quick and well-written reference book best suited for graduate students and researchers with some background in optimization. Since the chapters of the book are (almost) independent one can read them separately, too.