scientific article; zbMATH DE number 4023107
Iterative aggregation methods are used to solve large scale problems by decomposing the initial problem into a number of subproblems and iteratively coordinating their local solutions using an aggregate model. In this book the authors apply iterative aggregation models to systems of equations and to unconstrained and constrained optimization problems. In their study, they also show that many well-known procedures for unconstrained optimization problems can be treated as limiting cases of iterative aggregation algorithms. The book contains six chapters organized in three parts. In part I, general methods of solving extremum problems and systems of equations are reviewed in Chapter 1, and decomposition methods in solving such problems in Chapter 2. Part II contains two chapters on iterative aggregation methods in solving systems of equations. In Chapter 3, special interproduct input-output models are considered. After a discussion of the so called ``basic iterative aggregation methods, which are known to converge only in some special cases, some new algorithms, which are guaranteed to converge globally, are proposed. Chapter 4 deals with the general systems of equations, where two and one level iterative aggregation algorithms with minimization of imbalances, and algorithms for optimal interproduct input-output models are considered. The numerical experiments with the algorithms suggest that incorporating aggregation into a classical solution algorithm usually increases the rate of convergence by a factor of 1.5 to 2, and that algorithms based on successive approximations are superior to those based on coordinate descent methods. In Part III, iterative aggregation algorithms for optimization problems are considered. In Chapter 5, two and one level algorithms as well as algorithms based on gradient methods for unconstrained optimization problems are dealt with. Finally, Chapter 6 is devoted to constrained optimization problems, where - application of the iterative aggregation algorithms developed for unconstrained problems to constrained problems with variables aggregation, - one-sided iterative algorithms for convex programming problems with constraints aggregation, - two- and one-level iterative algorithms with both variables and constraints aggregation, - general schemes for two- and one-level iterative aggregation algorithms are considered. The book presents the iterative aggregation theory in a systematic and rigorous way, and should be very helpful to those dealing with mathematical methods of coordinating large scale problems.
- An iterative aggregation-disaggregation algorithm for solving linear equations
- Mathematical programming with iterative aggregation
- Estimating equilibrium probabilities for band diagonal Markov chains using aggregation and disaggregation techniques
- Aggregation and decomposition for multi-divisional linear programs
- A posteriori error bounds in linear programming aggregation
- Decomposition and iterative aggregation in hierarchical and decentralised planning structures
- Aggregation in large-scale optimization
- An aggregate and iterative disaggregate algorithm with proven optimality in machine learning
- Solving network design problems via iterative aggregation
- scientific article; zbMATH DE number 3864992 (Why is no real title available?)
- Iterative aggregation: convergence rate
- scientific article; zbMATH DE number 4068572 (Why is no real title available?)
- scientific article; zbMATH DE number 4102817 (Why is no real title available?)
- scientific article; zbMATH DE number 41630 (Why is no real title available?)
- Gradient schemes in iterative aggregation procedures for variable-aggregated lp-problems
This page was built for publication:
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3765615)