# Active Set Algorithm for Large-Scale Continuous Knapsack Problems with Application to Topology Optimization Problems - Mathematics > Optimization and Control

Active Set Algorithm for Large-Scale Continuous Knapsack Problems with Application to Topology Optimization Problems - Mathematics > Optimization and Control - Download this document for free, or read online. Document in PDF available to download.

Abstract: The structure of many real-world optimization problems includes minimizationof a nonlinear or quadratic functional subject to bound and singly linearconstraints in the form of either equality or bilateral inequality which arecommonly called as continuous knapsack problems. Since there are efficientmethods to solve large-scale bound constrained nonlinear programs, it isdesirable to adapt these methods to solve knapsack problems, while preservingtheir efficiency and convergence theories. The goal of this paper is tointroduce a general framework to extend a box-constrained optimization solverto solve knapsack problems. This framework includes two main ingredients whichare On methods; in terms of the computational cost and required memory; forthe projection onto the knapsack constrains and the null-space manipulation ofthe related linear constraint. The main focus of this work is on the extensionof Hager-Zhang active set algorithm SIAM J. Optim. 2006, pp. 526-557. Themain reasons for this choice was its promising efficiency in practice as wellas its excellent convergence theories e.g., superlinear local convergence ratewithout strict complementarity assumption. Moreover, this method does not useany explicit form of second order information and-or solution of linear systemsin the course of optimization which makes it an ideal choice for large-scaleproblems. Moreover, application of Birgin and Mart{\-\i}nez active setalgorithm Comp. Opt. Appl. 2002, pp. 101-125 for knapsack problems is alsobriefly commented. The feasibility of the presented algorithm is supported bynumerical results for topology optimization problems.

Author: ** Ruhollah Tavakoli**

Source: https://arxiv.org/